title
Stanford CS229: Machine Learning Course, Lecture 1 - Andrew Ng (Autumn 2018)

description
For more information about Stanford’s Artificial Intelligence professional and graduate programs, visit: https://stanford.io/2Ze53pq Listen to the first lecture in Andrew Ng's machine learning course. This course provides a broad introduction to machine learning and statistical pattern recognition. Learn about both supervised and unsupervised learning as well as learning theory, reinforcement learning and control. Explore recent applications of machine learning and design and develop algorithms for machines. Andrew Ng is an Adjunct Professor of Computer Science at Stanford University. View more about Andrew on his website: https://www.andrewng.org/ To follow along with the course schedule and syllabus, visit: http://cs229.stanford.edu/syllabus-autumn2018.html 0:00 Introduction 05:21 Teaching team introductions 06:42 Goals for the course and the state of machine learning across research and industry 10:09 Prerequisites for the course 11:53 Homework, and a note about the Stanford honor code 16:57 Overview of the class project 25:57 Questions #AndrewNg #machinelearning

detail
{'title': 'Stanford CS229: Machine Learning Course, Lecture 1 - Andrew Ng (Autumn 2018)', 'heatmap': [{'end': 3845.636, 'start': 3747.03, 'weight': 0.718}], 'summary': 'The stanford cs229 machine learning course lectures cover the rapid growth of ai and machine learning, industry applications, class projects, logistical changes, supervised learning, support vector machines for autonomous driving, systematic engineering principles, and unsupervised learning applications in various domains.', 'chapters': [{'end': 299.93, 'segs': [{'end': 37.382, 'src': 'embed', 'start': 4.235, 'weight': 3, 'content': [{'end': 6.057, 'text': 'Welcome to CST 29 Machine Learning.', 'start': 4.235, 'duration': 1.822}, {'end': 12.563, 'text': 'Uh, some of you know that this class I taught at Stanford for a long time and this is often the class that, um,', 'start': 6.457, 'duration': 6.106}, {'end': 17.287, 'text': "I most look forward to teaching each year, because this is where we've helped, I think,", 'start': 12.563, 'duration': 4.724}, {'end': 21.23, 'text': 'several generations of Stanford students become experts in machine learning, go out,', 'start': 17.287, 'duration': 3.943}, {'end': 27.095, 'text': "build many of their products and services and startups that I'm sure many of you I pray all of you are using.", 'start': 21.23, 'duration': 5.865}, {'end': 28.257, 'text': 'uh uh, today.', 'start': 27.095, 'duration': 1.162}, {'end': 37.382, 'text': 'Um, so what I want to do today was spend some time talking over, uh, logistics and then, uh, spend some time, you know,', 'start': 28.997, 'duration': 8.385}], 'summary': 'Cst 29 machine learning class at stanford has helped several generations of students become experts in ml and build products and services.', 'duration': 33.147, 'max_score': 4.235, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I4235.jpg'}, {'end': 128.947, 'src': 'embed', 'start': 90.09, 'weight': 0, 'content': [{'end': 94.893, 'text': 'go, go transform healthcare, or go transform transportation, or go build a self-driving car.', 'start': 90.09, 'duration': 4.803}, {'end': 100.642, 'text': "um, and do all of these things that, um, after this class, anything you'd be able to do.", 'start': 94.893, 'duration': 5.749}, {'end': 100.862, 'text': 'you know?', 'start': 100.642, 'duration': 0.22}, {'end': 108.792, 'text': 'the majority of students applying the demand for AI skills, the demand for machine learning skills, is so vast.', 'start': 102.728, 'duration': 6.064}, {'end': 109.793, 'text': 'I think you all know that.', 'start': 108.832, 'duration': 0.961}, {'end': 118.339, 'text': "Um, and I think it's because machine learning has advanced so rapidly in the last few years that there are so many opportunities, um,", 'start': 110.133, 'duration': 8.206}, {'end': 122.102, 'text': 'to apply learning algorithms right, both in industry as well as in academia.', 'start': 118.339, 'duration': 3.763}, {'end': 128.947, 'text': 'I think today we have, um, the English department professors trying to apply learning algorithms to understand history better.', 'start': 122.282, 'duration': 6.665}], 'summary': 'Ai and machine learning skills in high demand for various industries post-class.', 'duration': 38.857, 'max_score': 90.09, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I90090.jpg'}, {'end': 181.265, 'src': 'embed', 'start': 147.095, 'weight': 4, 'content': [{'end': 153.464, 'text': 'So I think that, um, uh uh, if you look at it on a on a factual basis,', 'start': 147.095, 'duration': 6.369}, {'end': 159.729, 'text': 'the number of people doing very valuable machine learning projects today is much greater than it was six months ago,', 'start': 153.464, 'duration': 6.265}, {'end': 162.191, 'text': 'and six months ago is much greater than it was 12 months ago.', 'start': 159.729, 'duration': 2.462}, {'end': 169.917, 'text': 'And the amount of value, the amount of exciting, meaningful work being done in machine learning is, is, is very strongly going up.', 'start': 162.751, 'duration': 7.166}, {'end': 181.265, 'text': 'Um and I think that, given the rise of you know the- the amount of data we have, as well as the new machine learning tools that we have, um,', 'start': 170.497, 'duration': 10.768}], 'summary': 'The number of people doing valuable machine learning projects is increasing over time, driven by the rise in available data and new tools.', 'duration': 34.17, 'max_score': 147.095, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I147095.jpg'}, {'end': 213.985, 'src': 'embed', 'start': 189.971, 'weight': 2, 'content': [{'end': 196.875, 'text': "Um So, just as maybe I don't know 20 years ago was a good time to start working on this Internet thing,", 'start': 189.971, 'duration': 6.904}, {'end': 201.858, 'text': 'and a lot of people that started working on the Internet, like 20 years ago, had fantastic careers.', 'start': 196.875, 'duration': 4.983}, {'end': 206.481, 'text': 'I think today is a wonderful time to jump to machine learning.', 'start': 202.298, 'duration': 4.183}, {'end': 213.985, 'text': 'uh, and-, and- and the number of- and the opportunities for you to do unique things that no one is- no one else is doing, right?', 'start': 206.481, 'duration': 7.504}], 'summary': 'Now is a great time to pursue machine learning with unique opportunities.', 'duration': 24.014, 'max_score': 189.971, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I189971.jpg'}], 'start': 4.235, 'title': 'Machine learning and ai advancement', 'summary': 'Provides an introduction to cst 29 machine learning and discusses the rapid growth of ai and machine learning, emphasizing the potential for individuals to become experts and the significant increase in valuable machine learning projects over the past few years.', 'chapters': [{'end': 41.444, 'start': 4.235, 'title': 'Cst 29: machine learning intro', 'summary': 'Introduces cst 29 machine learning, taught at stanford, aiming to turn students into machine learning experts who go on to build products and services.', 'duration': 37.209, 'highlights': ['The class has helped several generations of Stanford students become experts in machine learning, who have built products and services and startups.', 'The instructor looks forward to teaching this class each year.', 'The chapter includes an introduction to machine learning and discusses logistics.', 'The class has been taught at Stanford for a long time.']}, {'end': 299.93, 'start': 42.464, 'title': 'The rise of ai and machine learning', 'summary': 'Discusses the rapid advancement and growing opportunities in machine learning, highlighting the potential for individuals to become pioneers in various industries and emphasizes the high demand for ai and machine learning skills, with a significant increase in valuable machine learning projects over the past few years.', 'duration': 257.466, 'highlights': ['The demand for AI and machine learning skills is vast, with the majority of students applying for such skills.', 'The number of people doing valuable machine learning projects has significantly increased over the past few years.', 'The rise of machine learning has led to numerous opportunities to apply learning algorithms in both industry and academia.', 'The speaker indicates the significant growth in meaningful work being done in machine learning, with the amount of value and exciting work strongly increasing.', 'The speaker encourages individuals to seize the current opportunities in machine learning, comparing it to the rise of the Internet 20 years ago and emphasizing the potential for unique and pioneering work in this field.']}], 'duration': 295.695, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I4235.jpg', 'highlights': ['The demand for AI and machine learning skills is vast, with the majority of students applying for such skills.', 'The rise of machine learning has led to numerous opportunities to apply learning algorithms in both industry and academia.', 'The speaker encourages individuals to seize the current opportunities in machine learning, comparing it to the rise of the Internet 20 years ago and emphasizing the potential for unique and pioneering work in this field.', 'The class has helped several generations of Stanford students become experts in machine learning, who have built products and services and startups.', 'The number of people doing valuable machine learning projects has significantly increased over the past few years.']}, {'end': 1009.041, 'segs': [{'end': 381.818, 'src': 'embed', 'start': 347.834, 'weight': 1, 'content': [{'end': 354.558, 'text': 'Uh, and so bring a lot of um uh, technical experience, uh, technical experience in machine learning,', 'start': 347.834, 'duration': 6.724}, {'end': 357.52, 'text': 'as well as practical know-how on how to actually make these things work.', 'start': 354.558, 'duration': 2.962}, {'end': 361.523, 'text': 'And with the large class that we have, we have a large TA team.', 'start': 358.161, 'duration': 3.362}, {'end': 366.646, 'text': "Um, I- maybe I won't introduce all of the TAs here today, but you meet many of them throughout this course here.", 'start': 361.803, 'duration': 4.843}, {'end': 373.67, 'text': "But the TA's expertise span everything from computer vision to natural language processing, to computer technology, to robotics, and so um.", 'start': 366.666, 'duration': 7.004}, {'end': 381.818, 'text': 'through this quarter, as you work on your class projects, I hope that you get a lot of uh, help and advice and mentoring from the TAs.', 'start': 374.691, 'duration': 7.127}], 'summary': 'Technical experience in machine learning, large ta team spanning computer vision, nlp, robotics, providing help and mentoring.', 'duration': 33.984, 'max_score': 347.834, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I347834.jpg'}, {'end': 470.785, 'src': 'embed', 'start': 441.628, 'weight': 3, 'content': [{'end': 450.233, 'text': "Um, one of the things I find very exciting about machine learning is that it's no longer a sort of pure tech company only kind of thing, right?", 'start': 441.628, 'duration': 8.605}, {'end': 451.994, 'text': 'I think that many years ago,', 'start': 450.353, 'duration': 1.641}, {'end': 462.18, 'text': 'um machine learning it was like a thing that you know the computer science department would do and that the elite AI companies like Google and Facebook and Baidu and Microsoft would do.', 'start': 451.994, 'duration': 10.186}, {'end': 470.785, 'text': 'Uh, but now it is so pervasive that even companies that are not traditional tech companies see a huge need to apply these tools.', 'start': 462.86, 'duration': 7.925}], 'summary': 'Machine learning is no longer limited to tech companies, now pervasive across industries.', 'duration': 29.157, 'max_score': 441.628, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I441628.jpg'}, {'end': 507.49, 'src': 'embed', 'start': 477.608, 'weight': 0, 'content': [{'end': 484.672, 'text': 'I I led the Google Brain team, which helped Google transform from what was already a great company 10 years ago to today which is, you know,', 'start': 477.608, 'duration': 7.064}, {'end': 485.713, 'text': 'a great AI company.', 'start': 484.672, 'duration': 1.041}, {'end': 495.9, 'text': "And then I also led the AI group at Baidu and let the company's technology and strategy to help Baidu also transform from what was already a great company many years ago to today,", 'start': 485.773, 'duration': 10.127}, {'end': 498.702, 'text': "arguably China's greatest AI company.", 'start': 495.9, 'duration': 2.802}, {'end': 507.49, 'text': "So having built the teams that led the AI transformations of two large tech companies, I feel like that's a great thing to do.", 'start': 498.742, 'duration': 8.748}], 'summary': 'Led ai transformations at google and baidu, making them great ai companies.', 'duration': 29.882, 'max_score': 477.608, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I477608.jpg'}, {'end': 542.52, 'src': 'embed', 'start': 518.238, 'weight': 2, 'content': [{'end': 527.128, 'text': 'Um, but after this class, I hope that each one of you will be well qualified to get a job at a shiny tech company and do machine learning there,', 'start': 518.238, 'duration': 8.89}, {'end': 531.453, 'text': 'or go into one of these other industries and do very valuable machine learning projects there.', 'start': 527.128, 'duration': 4.325}, {'end': 542.52, 'text': 'Um. and, in addition, if any of you, um, are taking this class with the primary goal of uh being able to do research, uh, in machine learning,', 'start': 532.176, 'duration': 10.344}], 'summary': 'Equip students for tech jobs or valuable ml projects, and research prospects.', 'duration': 24.282, 'max_score': 518.238, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I518238.jpg'}, {'end': 590.129, 'src': 'embed', 'start': 565.189, 'weight': 4, 'content': [{'end': 577.84, 'text': "Um so today? uh so- so, just as machine learning is evolving rapidly um, the whole teaching team would've been, uh, constantly updating CS229 as well.", 'start': 565.189, 'duration': 12.651}, {'end': 579.961, 'text': 'So, um, it is actually very interesting.', 'start': 577.94, 'duration': 2.021}, {'end': 582.664, 'text': 'I feel like the pace of progress in machine learning has accelerated.', 'start': 580.021, 'duration': 2.643}, {'end': 590.129, 'text': 'So it- it actually feels like that, uh, the amount we changed the class year over year has been increasing over time.', 'start': 582.684, 'duration': 7.445}], 'summary': 'Machine learning progress accelerates, cs229 constantly updated.', 'duration': 24.94, 'max_score': 565.189, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I565189.jpg'}, {'end': 887.298, 'src': 'embed', 'start': 865.321, 'weight': 6, 'content': [{'end': 873.406, 'text': 'But um all the respect for the Stanford Honor Code, as well as for uh uh, you know for- for- for students kind of doing their own work.', 'start': 865.321, 'duration': 8.085}, {'end': 875.307, 'text': 'we ask you to basically do your own work.', 'start': 873.406, 'duration': 1.901}, {'end': 880.351, 'text': "uh, for the um, it's okay to discuss it, but after discussing homework problems with friends,", 'start': 875.847, 'duration': 4.504}, {'end': 887.298, 'text': 'ultimately we ask you to write up your problems by yourself so that the homework submissions re- reflect your own work, right?', 'start': 880.351, 'duration': 6.947}], 'summary': 'Respect the stanford honor code, do your own work, discuss but write up problems by yourself.', 'duration': 21.977, 'max_score': 865.321, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I865321.jpg'}], 'start': 301.03, 'title': 'Machine learning and its industry applications', 'summary': "Introduces the teaching team and the role of tas, emphasizing the expertise and support available to students, with a goal for students to become experts in machine learning and apply it to various fields after the course. it also discusses the increasing relevance of machine learning beyond traditional tech companies, the speaker's influential role in transforming google and baidu into prominent ai companies, the evolving nature of the cs229 class to keep up with accelerating progress in machine learning, and the emphasis on the stanford honor code for maintaining the integrity of completing cs229.", 'chapters': [{'end': 441.108, 'start': 301.03, 'title': 'Introduction to machine learning', 'summary': 'Introduces the teaching team and the role of tas, emphasizing the expertise and support available to students, with a goal for students to become experts in machine learning and apply it to various fields after the course.', 'duration': 140.078, 'highlights': ['The teaching team, including the class coordinator and co-head TAs, provide technical expertise and support to students, with a large TA team spanning various machine learning application areas.', 'The goal is for students to become experts in machine learning and apply it to academic and professional settings, such as mechanical engineering, electrical engineering, law, and education.', 'Students will receive mentoring and advice from TAs throughout the course, with a focus on matching students to TAs relevant to their class projects.']}, {'end': 1009.041, 'start': 441.628, 'title': 'Machine learning in industry', 'summary': "Discusses the increasing relevance of machine learning beyond traditional tech companies, the speaker's influential role in transforming google and baidu into prominent ai companies, the evolving nature of the cs229 class to keep up with accelerating progress in machine learning, and the emphasis on the stanford honor code for maintaining the integrity of completing cs229.", 'duration': 567.413, 'highlights': ['The speaker led the AI transformations of Google and Baidu, significantly impacting their status as prominent AI companies today.', 'The relevance of machine learning has expanded to non-tech companies, creating a significant need to apply machine learning tools.', 'The CS229 class has evolved to keep up with the accelerating progress in the field of machine learning, necessitating constant updates and logistical changes.', 'The emphasis on the Stanford Honor Code for maintaining the integrity of completing CS229, with a specific focus on individual work for homework submissions.']}], 'duration': 708.011, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I301030.jpg', 'highlights': ['The speaker led the AI transformations of Google and Baidu, significantly impacting their status as prominent AI companies today.', 'The teaching team, including the class coordinator and co-head TAs, provide technical expertise and support to students, with a large TA team spanning various machine learning application areas.', 'The goal is for students to become experts in machine learning and apply it to academic and professional settings, such as mechanical engineering, electrical engineering, law, and education.', 'The relevance of machine learning has expanded to non-tech companies, creating a significant need to apply machine learning tools.', 'The CS229 class has evolved to keep up with the accelerating progress in the field of machine learning, necessitating constant updates and logistical changes.', 'Students will receive mentoring and advice from TAs throughout the course, with a focus on matching students to TAs relevant to their class projects.', 'The emphasis on the Stanford Honor Code for maintaining the integrity of completing CS229, with a specific focus on individual work for homework submissions.']}, {'end': 1480.319, 'segs': [{'end': 1032.776, 'src': 'embed', 'start': 1009.501, 'weight': 0, 'content': [{'end': 1016.687, 'text': 'Um, one of the best parts of, um, the class is, um, the class project.', 'start': 1009.501, 'duration': 7.186}, {'end': 1023.512, 'text': 'Um, and so, you know, one of the goals of the class is to leave you well-qualified to do a meaningful machine learning project.', 'start': 1017.107, 'duration': 6.405}, {'end': 1032.776, 'text': 'And so, uh, one of the best ways to make sure you have that skill set is through this class and hopefully with the help of some of our TAs.', 'start': 1023.872, 'duration': 8.904}], 'summary': 'Class aims to prepare for meaningful machine learning project with the help of tas.', 'duration': 23.275, 'max_score': 1009.501, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1009501.jpg'}, {'end': 1083.316, 'src': 'embed', 'start': 1056.605, 'weight': 1, 'content': [{'end': 1064.787, 'text': 'pick an application that excites you and to apply machine learning to it and see if you can build a good machine learning system for some application area.', 'start': 1056.605, 'duration': 8.182}, {'end': 1071.609, 'text': "And so, um, if you go to the course website, you know cs229.stanford.edu, and look at previous year's projects,", 'start': 1065.307, 'duration': 6.302}, {'end': 1074.63, 'text': 'you- you- you see machine learning projects applied to pretty much.', 'start': 1071.609, 'duration': 3.021}, {'end': 1078.011, 'text': 'you know pretty much every imaginable application under the sun.', 'start': 1074.63, 'duration': 3.381}, {'end': 1083.316, 'text': "everything from, I don't know, diagnosing cancer to creating art, to uh.", 'start': 1078.371, 'duration': 4.945}], 'summary': 'Apply machine learning to diverse applications for building effective systems.', 'duration': 26.711, 'max_score': 1056.605, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1056605.jpg'}, {'end': 1157.852, 'src': 'embed', 'start': 1132.868, 'weight': 3, 'content': [{'end': 1138.194, 'text': 'And so, um, after class, today also encourage you to start making friends in the class,', 'start': 1132.868, 'duration': 5.326}, {'end': 1144.461, 'text': 'both for the purpose of forming study groups as well as for the purpose of maybe finding a small group to do a class project with.', 'start': 1138.194, 'duration': 6.267}, {'end': 1149.726, 'text': 'Uh, we ask you to form project groups of, um, up to size three.', 'start': 1145.622, 'duration': 4.104}, {'end': 1153.369, 'text': 'Uh, uh, most project groups end up being size two or three.', 'start': 1149.746, 'duration': 3.623}, {'end': 1157.852, 'text': "Um, if you insist on doing it by yourself, right, without any partners, that's actually okay too.", 'start': 1153.389, 'duration': 4.463}], 'summary': 'Encouraging students to form study groups and project teams, with project groups typically being size two or three.', 'duration': 24.984, 'max_score': 1132.868, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1132868.jpg'}, {'end': 1291.932, 'src': 'embed', 'start': 1266.431, 'weight': 5, 'content': [{'end': 1271.054, 'text': 'uh on- held on Fridays that are- and everything we do, including the- see all the-.', 'start': 1266.431, 'duration': 4.623}, {'end': 1276.878, 'text': 'all the lectures and discussion sections are recorded and broadcast through SCPD, uh, through the online website.', 'start': 1271.054, 'duration': 5.824}, {'end': 1288.369, 'text': 'Um, and one of- and, uh, discussion sections, uh, are taught, uh, usually by the TAs on Fridays, and attendance at discussion sections is optional.', 'start': 1277.538, 'duration': 10.831}, {'end': 1291.932, 'text': 'Uh and what- what I mean is that, um, you, you know,', 'start': 1288.669, 'duration': 3.263}], 'summary': 'Lectures and discussion sections are recorded and broadcast through scpd, with optional attendance at discussion sections.', 'duration': 25.501, 'max_score': 1266.431, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1266431.jpg'}, {'end': 1424.19, 'src': 'embed', 'start': 1400.221, 'weight': 4, 'content': [{'end': 1407.964, 'text': "but um definitely encourage you to uh participate actively on Piazza and also to answer other students' questions.", 'start': 1400.221, 'duration': 7.743}, {'end': 1413.445, 'text': 'I think that one of the best ways to learn uh, as well as contribute, you know, back to the class as a whole,', 'start': 1408.324, 'duration': 5.121}, {'end': 1416.306, 'text': 'is if you see someone else ask a question on Piazza.', 'start': 1413.445, 'duration': 2.861}, {'end': 1421.529, 'text': 'if you jump in and help answer that uh that- that often helps you and helps your classmates.', 'start': 1416.306, 'duration': 5.223}, {'end': 1424.19, 'text': 'So I strongly encourage you to do that.', 'start': 1421.549, 'duration': 2.641}], 'summary': 'Active participation on piazza benefits learning and class engagement.', 'duration': 23.969, 'max_score': 1400.221, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1400221.jpg'}], 'start': 1009.501, 'title': 'Importance of class projects in machine learning', 'summary': 'Emphasizes the significance of the class project in cs229, aiming to equip students with the skills to undertake a meaningful machine learning project and encourages group work with ta support. it also highlights the importance of drawing inspiration from previous projects, forming small groups, attending discussion sections, and participating in online discussions to enhance learning.', 'chapters': [{'end': 1093.986, 'start': 1009.501, 'title': 'Importance of class projects in machine learning', 'summary': 'Emphasizes the significance of the class project in cs229, aiming to equip students with the skills to undertake a meaningful machine learning project, with a focus on applying machine learning to various real-world applications. the class encourages students to work in small groups and provides support through tas.', 'duration': 84.485, 'highlights': ['The class project is a key component, aiming to leave students well-qualified to undertake a meaningful machine learning project, supported by TAs.', 'Students are encouraged to brainstorm and collaborate with peers to select an application area that excites them and apply machine learning to build a system for it.', "The course website showcases previous year's projects, demonstrating the application of machine learning to a wide range of areas, including diagnosing cancer, creating art, and engineering applications."]}, {'end': 1480.319, 'start': 1093.986, 'title': 'Literature application and logistics', 'summary': 'Emphasizes the importance of drawing inspiration from previous projects, forming small project groups, attending discussion sections, and actively participating in the online discussion board piazza to enhance learning and contribute to the class as a whole.', 'duration': 386.333, 'highlights': ['The chapter emphasizes drawing inspiration from previous projects and forming small project groups of up to three members, with the majority being size two or three.', 'The discussion sections are recorded and broadcast through SCPD, with attendance being optional and used to cover prerequisite material in greater depth for the first few weeks.', "Active participation on the online discussion board Piazza is strongly encouraged, with students urged to answer each other's questions and use the platform for technical and logistical queries."]}], 'duration': 470.818, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1009501.jpg', 'highlights': ['The class project is a key component, aiming to leave students well-qualified to undertake a meaningful machine learning project, supported by TAs.', 'Students are encouraged to brainstorm and collaborate with peers to select an application area that excites them and apply machine learning to build a system for it.', "The course website showcases previous year's projects, demonstrating the application of machine learning to a wide range of areas, including diagnosing cancer, creating art, and engineering applications.", 'The chapter emphasizes drawing inspiration from previous projects and forming small project groups of up to three members, with the majority being size two or three.', "Active participation on the online discussion board Piazza is strongly encouraged, with students urged to answer each other's questions and use the platform for technical and logistical queries.", 'The discussion sections are recorded and broadcast through SCPD, with attendance being optional and used to cover prerequisite material in greater depth for the first few weeks.']}, {'end': 2334.404, 'segs': [{'end': 1528.402, 'src': 'embed', 'start': 1481.22, 'weight': 0, 'content': [{'end': 1485.784, 'text': "Um, and we'll be using Gradescope as well, um, uh, to- for- for online grading.", 'start': 1481.22, 'duration': 4.564}, {'end': 1488.586, 'text': "And- and if you- if you don't know what Gradescope is, don't worry about it.", 'start': 1485.984, 'duration': 2.602}, {'end': 1492.59, 'text': "We'll- we'll- we'll send you links and show you how to use it later.", 'start': 1488.666, 'duration': 3.924}, {'end': 1506.586, 'text': 'Um, oh and uh, uh again relative to one- one last logistical thing to plan for um, unlike, uh, previous, um, uh, years where we taught CS239, uh.', 'start': 1493.837, 'duration': 12.749}, {'end': 1513.831, 'text': "so we're constantly updating the syllabus right the technical content, to try to show you the latest machine learning algorithms, uh and uh, uh,", 'start': 1506.586, 'duration': 7.245}, {'end': 1516.853, 'text': "the two big little changes we're making this year, I guess.", 'start': 1513.831, 'duration': 3.022}, {'end': 1523.438, 'text': 'One is uh, Python instead of MATLAB, and the other one is uh um instead of having a uh.', 'start': 1517.013, 'duration': 6.425}, {'end': 1524.639, 'text': 'uh. midterm exam.', 'start': 1523.438, 'duration': 1.201}, {'end': 1527.061, 'text': "you know that's a timed midterm.", 'start': 1524.639, 'duration': 2.422}, {'end': 1528.402, 'text': "uh, we're planning to, uh.", 'start': 1527.061, 'duration': 1.341}], 'summary': 'Using gradescope for online grading, updating syllabus with latest ml algorithms, switching from matlab to python, and replacing timed midterm with alternative assessment methods.', 'duration': 47.182, 'max_score': 1481.22, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1481220.jpg'}, {'end': 1636.921, 'src': 'embed', 'start': 1585.304, 'weight': 6, 'content': [{'end': 1594.329, 'text': 'So someone else is teaching it in spring quarter and, um, uh, I actually did not know it was gonna be offered in winter.', 'start': 1585.304, 'duration': 9.025}, {'end': 1599.052, 'text': "Yeah I don't believe it's offered in winter.", 'start': 1595.89, 'duration': 3.162}, {'end': 1599.572, 'text': 'Yeah Right.', 'start': 1599.072, 'duration': 0.5}, {'end': 1599.752, 'text': 'Yeah Yes.', 'start': 1599.592, 'duration': 0.16}, {'end': 1607.443, 'text': 'Yeah, So, so I think a great guy and teaching it uh in.', 'start': 1600.713, 'duration': 6.73}, {'end': 1613.734, 'text': "sorry, Camille, if I'm saying it right, uh, teaching it in uh, spring, uh, uh, and I don't think it's offered in winter.", 'start': 1607.443, 'duration': 6.291}, {'end': 1615.978, 'text': 'Anything else? Go ahead.', 'start': 1615.016, 'duration': 0.962}, {'end': 1620.717, 'text': 'Uh, will this one section be recorded? Yes, it will be.', 'start': 1618.777, 'duration': 1.94}, {'end': 1626.399, 'text': "Oh, and by the way, if- if- if you wonder why I'm recording the- I'm repeating the question, I know it feels weird.", 'start': 1621.117, 'duration': 5.282}, {'end': 1630.539, 'text': "I'm recording for the microphone so that- so that people watching this at home can hear the question.", 'start': 1626.419, 'duration': 4.12}, {'end': 1636.921, 'text': 'But, uh, both the lectures and the discussion sections, uh, will be- will be recorded and put on the website.', 'start': 1630.579, 'duration': 6.342}], 'summary': 'Course will be taught in spring, lectures and sections will be recorded.', 'duration': 51.617, 'max_score': 1585.304, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1585304.jpg'}, {'end': 1691.197, 'src': 'embed', 'start': 1658.336, 'weight': 4, 'content': [{'end': 1662.158, 'text': 'In previous years, one of the feedback we got was that the office hours are really crowded.', 'start': 1658.336, 'duration': 3.822}, {'end': 1667.24, 'text': 'So- so we have 60- 60 hours of- 60 office hour slots per week this year.', 'start': 1662.218, 'duration': 5.022}, {'end': 1668.36, 'text': 'That- that seems like a lot.', 'start': 1667.5, 'duration': 0.86}, {'end': 1674.443, 'text': "So hopefully, if you need to track down one of us, track down the tier to get help, hopefully that'll- that'll make it easier for you to do so.", 'start': 1668.38, 'duration': 6.063}, {'end': 1674.823, 'text': 'Go ahead.', 'start': 1674.603, 'duration': 0.22}, {'end': 1691.197, 'text': 'Say that again? Will- Oh, will logistical things like when homework is due be covered in lectures? Uh, we have, um, yes.', 'start': 1678.844, 'duration': 12.353}], 'summary': 'Increased office hours to 60 slots per week to address crowded feedback.', 'duration': 32.861, 'max_score': 1658.336, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1658336.jpg'}, {'end': 1806.266, 'src': 'embed', 'start': 1771.485, 'weight': 2, 'content': [{'end': 1773.726, 'text': "So- so please don't all sign up for CS239A.", 'start': 1771.485, 'duration': 2.241}, {'end': 1778.589, 'text': 'because, um, uh, we- 239A does not have the capacity this quarter.', 'start': 1774.787, 'duration': 3.802}, {'end': 1789.816, 'text': 'But, uh, CS 239A is a, um, much less mathematical and, uh, much more applied, uh, uh, uh, a relatively more applied version of machine learning.', 'start': 1779.07, 'duration': 10.746}, {'end': 1795.679, 'text': "And, uh, so I- I guess I'm teaching CS239A, CS230, and CS239 this quarter.", 'start': 1790.496, 'duration': 5.183}, {'end': 1799.022, 'text': 'Of the three, CS239 is the most mathematical.', 'start': 1795.84, 'duration': 3.182}, {'end': 1806.266, 'text': 'Um, it is a little bit less applied than CS239A, which is more applied machine learning, and CS230, which is deep learning.', 'start': 1799.722, 'duration': 6.544}], 'summary': 'Cs239a not available this quarter. cs239a is more applied machine learning. cs230 is deep learning.', 'duration': 34.781, 'max_score': 1771.485, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1771485.jpg'}, {'end': 1899.646, 'src': 'embed', 'start': 1874.636, 'weight': 3, 'content': [{'end': 1881.863, 'text': 'uh, and CS230 is more focused on deep learning algorithms, specifically right, which is a much narrow set of algorithms, but it is, you know,', 'start': 1874.636, 'duration': 7.227}, {'end': 1884.161, 'text': 'one of the hottest areas of deep learning.', 'start': 1881.863, 'duration': 2.298}, {'end': 1888.583, 'text': 'Uh, there is not that much overlap in content between the three classes.', 'start': 1884.541, 'duration': 4.042}, {'end': 1893.324, 'text': 'So if you actually take all three, you learn relatively different things from all of them.', 'start': 1888.783, 'duration': 4.541}, {'end': 1899.646, 'text': "Uh, in the past, we've had students simultaneously take 229 and 229A, and there is a little bit of overlap.", 'start': 1893.344, 'duration': 6.302}], 'summary': 'Cs230 focuses on narrow deep learning algorithms, one of the hottest areas in deep learning, with little overlap in content compared to other classes.', 'duration': 25.01, 'max_score': 1874.636, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1874636.jpg'}, {'end': 2019.007, 'src': 'embed', 'start': 1990.007, 'weight': 5, 'content': [{'end': 1991.288, 'text': 'And machine learning is bigger than deep learning.', 'start': 1990.007, 'duration': 1.281}, {'end': 1999.352, 'text': 'Um, one of the great things about being a Stanford student is you can, and I think should, take multiple classes right?', 'start': 1992.009, 'duration': 7.343}, {'end': 2004.535, 'text': 'I think that you know CS329 has for many years been the core of the machine learning world at Stanford.', 'start': 1999.512, 'duration': 5.023}, {'end': 2012.061, 'text': 'uh, but even beyond, CS, 239 is worth your while to take multiple classes and gain multiple perspectives.', 'start': 2004.535, 'duration': 7.526}, {'end': 2019.007, 'text': 'So- so if you want to, uh, be really effective, you know, after you graduate from Stanford, you do want to be an expert in machine learning.', 'start': 2012.101, 'duration': 6.906}], 'summary': 'Stanford emphasizes the importance of taking multiple machine learning classes for expertise.', 'duration': 29, 'max_score': 1990.007, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1990007.jpg'}, {'end': 2126.476, 'src': 'embed', 'start': 2100.2, 'weight': 7, 'content': [{'end': 2104.562, 'text': 'But the thing I find even more exciting is is the meaningful work we could do, right?', 'start': 2100.2, 'duration': 4.362}, {'end': 2105.942, 'text': 'I think that You know.', 'start': 2104.602, 'duration': 1.34}, {'end': 2111.166, 'text': "I think that every time there's a major technological disruption, which there is now through machine learning, um,", 'start': 2105.942, 'duration': 5.224}, {'end': 2113.928, 'text': 'it gives us an opportunity to remake large parts of the world.', 'start': 2111.166, 'duration': 2.762}, {'end': 2122.193, 'text': "And if we behave ethically, in a principle way and use these superpowers of machine learning to do things that you know helps people's lives right?", 'start': 2114.528, 'duration': 7.665}, {'end': 2126.476, 'text': 'Maybe we could, um, uh, maybe you could improve the healthcare system.', 'start': 2122.253, 'duration': 4.223}], 'summary': 'Machine learning offers opportunity to remake world ethically, improving healthcare system.', 'duration': 26.276, 'max_score': 2100.2, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I2100200.jpg'}], 'start': 1481.22, 'title': 'Logistical changes and machine learning at stanford', 'summary': 'Covers logistical changes for cs239, including use of gradescope, syllabus and content updates, and switch to python. it also discusses machine learning classes at stanford, emphasizing potential impact on the world and importance of taking multiple classes.', 'chapters': [{'end': 1615.978, 'start': 1481.22, 'title': 'Logistical changes and updates for cs239', 'summary': 'Covers the logistical changes for cs239, including the use of gradescope for online grading, updates to the syllabus and technical content, and the switch to python instead of matlab, as well as the replacement of timed midterm with a take-home midterm.', 'duration': 134.758, 'highlights': ['The course will be using Gradescope for online grading and will be updating the syllabus and technical content to include the latest machine learning algorithms.', 'The switch to Python instead of MATLAB and the replacement of a timed midterm with a take-home midterm are the two significant changes being made for the course.', 'The course is offered every quarter, with different instructors for each rendition, and it is confirmed to be offered in the spring quarter, but not in the winter.', 'The use of Gradescope for online grading will be explained later, and details about the take-home midterm will be provided as well.', 'The instructor mentions that the course will be constantly updating the syllabus and technical content to include the latest machine learning algorithms.']}, {'end': 2334.404, 'start': 1618.777, 'title': 'Overview of machine learning at stanford', 'summary': 'Discusses the recording of lectures and office hours, differences between various machine learning classes, the importance of taking multiple classes, and the meaning and definitions of machine learning, emphasizing its potential to make a meaningful impact on the world.', 'duration': 715.627, 'highlights': ['The lectures and the discussion sections will be recorded and put on the website, and there will be 60 office hour slots per week to improve accessibility for students.', 'CS239A is a relatively more applied version of machine learning with capped enrollment, and CS230 focuses more on deep learning algorithms.', 'The speaker emphasizes the importance of taking multiple classes and gaining multiple perspectives in order to become an expert in machine learning and deep learning.', 'The speaker shares personal views on the meaningful work that can be achieved through machine learning and presents different definitions, highlighting its potential to reshape the world in a positive way.']}], 'duration': 853.184, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I1481220.jpg', 'highlights': ['The course will be using Gradescope for online grading and updating syllabus and content.', 'The switch to Python and take-home midterm are significant changes for the course.', 'CS239A is a more applied version of machine learning with capped enrollment.', 'CS230 focuses more on deep learning algorithms.', '60 office hour slots per week will be available to improve accessibility for students.', 'Importance of taking multiple classes and gaining multiple perspectives is emphasized.', 'The course is confirmed to be offered in the spring quarter, but not in the winter.', 'The potential impact of machine learning on the world is highlighted.', 'The course will be constantly updating the syllabus and technical content.', 'Lectures and discussion sections will be recorded and put on the website.']}, {'end': 2975.999, 'segs': [{'end': 2481.399, 'src': 'embed', 'start': 2447.709, 'weight': 0, 'content': [{'end': 2451.412, 'text': 'and then the vertical axis will plot the price of the house right?', 'start': 2447.709, 'duration': 3.703}, {'end': 2458.248, 'text': 'And um, maybe a dataset looks like that.', 'start': 2451.952, 'duration': 6.296}, {'end': 2466.012, 'text': "Um, and so horizontal axis, I guess we'll call this X, and vertical axis, we'll call that Y.", 'start': 2459.009, 'duration': 7.003}, {'end': 2473.735, 'text': 'So, um, the supervised learning problem is, given the data set like this, to find the relationship mapping from X to Y, right?', 'start': 2466.012, 'duration': 7.723}, {'end': 2481.399, 'text': "And so um, for example, let's say you- let's say uh, let's say you have a-, let's say you're fortunate enough to own a house in Palo Alto, right?", 'start': 2474.015, 'duration': 7.384}], 'summary': 'Supervised learning involves finding the relationship mapping from x to y in a dataset, for example, predicting house prices in palo alto.', 'duration': 33.69, 'max_score': 2447.709, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I2447709.jpg'}, {'end': 2575.932, 'src': 'embed', 'start': 2540.585, 'weight': 1, 'content': [{'end': 2543.367, 'text': 'maybe one of the simplest learning algorithms.', 'start': 2540.585, 'duration': 2.782}, {'end': 2548.17, 'text': 'Um. given the data set like this, there are many possible ways to learn the mapping,', 'start': 2544.028, 'duration': 4.142}, {'end': 2552.293, 'text': 'to learn the function mapping from the input size to the estimated price.', 'start': 2548.17, 'duration': 4.123}, {'end': 2556.076, 'text': 'And so, um, maybe you want to fit a quadratic function instead.', 'start': 2552.633, 'duration': 3.443}, {'end': 2558.197, 'text': 'Maybe that actually fits the data a little bit better.', 'start': 2556.096, 'duration': 2.101}, {'end': 2560.979, 'text': 'And so how do you choose among different models will be?', 'start': 2558.377, 'duration': 2.602}, {'end': 2566.623, 'text': "uh, either automatically or manual intervention will be- will be something we'll spend a lot of time talking about.", 'start': 2560.979, 'duration': 5.644}, {'end': 2575.932, 'text': 'Now, to give a little bit more, um, to define a few more things, this particular example is a problem called a regression problem.', 'start': 2567.782, 'duration': 8.15}], 'summary': 'Discussing learning algorithms and model selection for a regression problem.', 'duration': 35.347, 'max_score': 2540.585, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I2540585.jpg'}, {'end': 2613.98, 'src': 'embed', 'start': 2585.532, 'weight': 2, 'content': [{'end': 2589.713, 'text': "Um, in contrast, here is a- here's a different type of problem.", 'start': 2585.532, 'duration': 4.181}, {'end': 2598.595, 'text': "Um. so problem that some of my friends were working on and- and I'll simplify it was- was a healthcare problem where, uh, they were looking at, uh,", 'start': 2590.693, 'duration': 7.902}, {'end': 2604.996, 'text': 'breast cancer, uh, breast tumors, um, and trying to decide if a tumor is benign or malignant.', 'start': 2598.595, 'duration': 6.401}, {'end': 2608.898, 'text': "Right. So a tumor, you know, sort of a lump in a- in a woman's breast.", 'start': 2605.276, 'duration': 3.622}, {'end': 2613.98, 'text': 'um is, uh, uh, can be malign or cancerous.', 'start': 2608.898, 'duration': 5.082}], 'summary': 'Healthcare problem: distinguishing benign from malignant breast tumors.', 'duration': 28.448, 'max_score': 2585.532, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I2585532.jpg'}, {'end': 2827.03, 'src': 'embed', 'start': 2801.909, 'weight': 3, 'content': [{'end': 2810.036, 'text': 'Right So, um, it turns out that, uh, uh, in both of these examples, the input x was one-dimensional.', 'start': 2801.909, 'duration': 8.127}, {'end': 2811.237, 'text': 'It was a single row number.', 'start': 2810.076, 'duration': 1.161}, {'end': 2817.342, 'text': 'For most of the, um, machine learning applications you work with, the input x will be multi-dimensional.', 'start': 2811.857, 'duration': 5.485}, {'end': 2821.085, 'text': "You won't be given just one number and also predict another number.", 'start': 2817.382, 'duration': 3.703}, {'end': 2827.03, 'text': 'Instead, you often be given, uh, multiple features or multiple numbers to predict another number.', 'start': 2821.646, 'duration': 5.384}], 'summary': 'In machine learning, inputs are often multi-dimensional, not just single numbers.', 'duration': 25.121, 'max_score': 2801.909, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I2801909.jpg'}], 'start': 2334.805, 'title': 'Supervised learning in machine learning', 'summary': 'Introduces the concept of supervised learning, illustrating with a housing price prediction example and emphasizes the importance of choosing the right model. it also discusses a healthcare problem of classifying breast tumors as benign or malignant, addressing the challenge of working with multi-dimensional input features.', 'chapters': [{'end': 2584.911, 'start': 2334.805, 'title': 'Supervised learning in machine learning', 'summary': 'Introduces the concept of supervised learning in machine learning, where the goal is to learn a mapping from input data to labels, illustrated through an example of predicting housing prices based on house size, emphasizing the importance of choosing the right model for accurate prediction.', 'duration': 250.106, 'highlights': ['The chapter explains the concept of supervised learning, which involves learning a mapping from input data to labels, exemplified through the task of predicting housing prices based on house size.', 'It emphasizes the importance of choosing the right model for learning the mapping, such as fitting a straight line or a quadratic function to the data, to achieve accurate predictions.', 'Further discussions include the classification of the example as a regression problem, where the value to be predicted (housing prices) is continuous.']}, {'end': 2975.999, 'start': 2585.532, 'title': 'Classification in healthcare', 'summary': 'Discusses a healthcare problem of classifying breast tumors as benign or malignant, using a dataset to learn a mapping from tumor size to malignancy, visualizing the data on a line, and addressing the challenge of working with multi-dimensional input features in machine learning applications.', 'duration': 390.467, 'highlights': ['The chapter discusses a healthcare problem of classifying breast tumors as benign or malignant.', 'Using a dataset to learn a mapping from tumor size to malignancy.', 'Visualizing the data on a line and addressing the challenge of working with multi-dimensional input features in machine learning applications.']}], 'duration': 641.194, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I2334805.jpg', 'highlights': ['The chapter explains supervised learning through housing price prediction.', 'Emphasizes the importance of choosing the right model for accurate predictions.', 'Discusses classifying breast tumors as benign or malignant in healthcare.', 'Addresses the challenge of working with multi-dimensional input features.']}, {'end': 3621.118, 'segs': [{'end': 3001.337, 'src': 'embed', 'start': 2975.999, 'weight': 2, 'content': [{'end': 2982.803, 'text': "if you're actually doing this in a- in a actual medical application there's a good chance that you'll be using a lot more features than just two.", 'start': 2975.999, 'duration': 6.804}, {'end': 2986.705, 'text': "Uh, and this means that you actually can't plot this data, right? It's too high dimensional.", 'start': 2983.143, 'duration': 3.562}, {'end': 2990.186, 'text': "You can't plot things higher than three-dimensional, or maybe four-dimensional, or something right?", 'start': 2986.725, 'duration': 3.461}, {'end': 2994.208, 'text': "And so when you have a lot of features, it's actually difficult to plot this data.", 'start': 2990.246, 'duration': 3.962}, {'end': 2996.449, 'text': "I'll come back to this in a second in learning theory.", 'start': 2994.248, 'duration': 2.201}, {'end': 3001.337, 'text': 'Um And one of the things you learn about.', 'start': 2997.07, 'duration': 4.267}], 'summary': 'In medical applications, using many features makes plotting high-dimensional data impossible.', 'duration': 25.338, 'max_score': 2975.999, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I2975999.jpg'}, {'end': 3037.365, 'src': 'embed', 'start': 3011.711, 'weight': 0, 'content': [{'end': 3019.415, 'text': 'One of the uh most fascinating results you learn is that, um, you also learn about an algorithm called the support vector machine,', 'start': 3011.711, 'duration': 7.704}, {'end': 3029.44, 'text': 'which uses not 1 or 2, or 3, or 10 or 100 or a million input features, but uses an infinite number of input features.', 'start': 3019.415, 'duration': 10.025}, {'end': 3037.365, 'text': 'Right? And so- so- so just be clear, if in this example, the state of a patient were represented as one number, you know, tumor size.', 'start': 3030.04, 'duration': 7.325}], 'summary': 'Support vector machine uses an infinite number of input features, such as tumor size.', 'duration': 25.654, 'max_score': 3011.711, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3011711.jpg'}, {'end': 3156.355, 'src': 'embed', 'start': 3126.525, 'weight': 1, 'content': [{'end': 3135.848, 'text': 'But at the heart of supervised learning is the idea that during training, uh, you are given inputs x together with the labels y,', 'start': 3126.525, 'duration': 9.323}, {'end': 3137.268, 'text': 'and you give it both at the same time.', 'start': 3135.848, 'duration': 1.42}, {'end': 3147.151, 'text': 'And the job of your learning algorithm is to, uh, find a mapping so that given a new x, you can map it to the most appropriate output y.', 'start': 3138.008, 'duration': 9.143}, {'end': 3156.355, 'text': 'Um, so this is a very old video, uh, made by, um, uh, Dean Pomelo, known for a long time as well, on using supervised learning for autonomous driving.', 'start': 3147.931, 'duration': 8.424}], 'summary': 'Supervised learning trains algorithm to map inputs to appropriate outputs.', 'duration': 29.83, 'max_score': 3126.525, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3126525.jpg'}, {'end': 3345.211, 'src': 'embed', 'start': 3288.49, 'weight': 3, 'content': [{'end': 3295.631, 'text': 'Um and as the algorithm learns using the back propagation learning algorithm or gradient descent which you learn about.', 'start': 3288.49, 'duration': 7.141}, {'end': 3297.732, 'text': 'uh, you actually learn about gradient descent this Wednesday.', 'start': 3295.631, 'duration': 2.101}, {'end': 3308.334, 'text': 'Um, you see that the neural networks outputs becomes less and less of this white smear, this white blur, but starts to become sharper, um,', 'start': 3298.572, 'duration': 9.762}, {'end': 3314.575, 'text': 'and starts to mimic more accurately the human selected driving direction.', 'start': 3308.334, 'duration': 6.241}, {'end': 3327.024, 'text': 'So this, um, is an example of supervised learning because the human driver demonstrates inputs x and outputs y.', 'start': 3317.501, 'duration': 9.523}, {'end': 3330.386, 'text': 'Uh, namely, if you see this in front of the car, steer like that.', 'start': 3327.024, 'duration': 3.362}, {'end': 3332.346, 'text': "So that's x and y.", 'start': 3330.566, 'duration': 1.78}, {'end': 3336.908, 'text': 'And, uh, after the learning algorithm has learned, um, you can.', 'start': 3332.346, 'duration': 4.562}, {'end': 3345.211, 'text': "then, uh well, he pushes a button, takes a hand off the steering wheel, um, and then it's using this neural network.", 'start': 3336.908, 'duration': 8.303}], 'summary': 'Neural network improves to mimic human driving direction, a supervised learning example.', 'duration': 56.721, 'max_score': 3288.49, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3288490.jpg'}, {'end': 3596.017, 'src': 'embed', 'start': 3572.522, 'weight': 4, 'content': [{'end': 3579.486, 'text': "and sometimes you could look at what they're doing and go yeah, you know, I could have told you six months ago that this approach is never gonna work,", 'start': 3572.522, 'duration': 6.964}, {'end': 3590.033, 'text': 'right?. Um and um, what I find is that the most skilled machine learning practitioners are very strategic, by which I mean that your skill at deciding.', 'start': 3579.486, 'duration': 10.547}, {'end': 3594.456, 'text': 'um, when you work on a machine learning project, you all- you- you have a lot of decisions to make, right?', 'start': 3590.033, 'duration': 4.423}, {'end': 3596.017, 'text': 'Do you collect more data??', 'start': 3594.836, 'duration': 1.181}], 'summary': 'Skilled machine learning practitioners are strategic in making decisions, such as data collection.', 'duration': 23.495, 'max_score': 3572.522, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3572522.jpg'}], 'start': 2975.999, 'title': 'Support vector machine and supervised learning for autonomous driving', 'summary': 'Discusses challenges of plotting high-dimensional data, introduces support vector machine algorithm handling infinite input features and explains supervised learning for autonomous driving using artificial neural network with training data from human drivers, emphasizing the importance of strategic decision-making in machine learning projects.', 'chapters': [{'end': 3147.151, 'start': 2975.999, 'title': 'Support vector machine and infinite dimensional vectors', 'summary': 'Discusses the challenges of plotting high-dimensional data and introduces the support vector machine algorithm, which can handle an infinite number of input features and is an effective learning algorithm for some problems.', 'duration': 171.152, 'highlights': ['The support vector machine algorithm can handle an infinite number of input features, providing a relatively effective learning algorithm for some problems.', 'It is challenging to plot high-dimensional data with many features, making it difficult to visualize and analyze the data.', 'Supervised learning involves training with inputs x and labels y to find a mapping for new x to the most appropriate output y, forming the basis of learning algorithms.']}, {'end': 3621.118, 'start': 3147.931, 'title': 'Supervised learning for autonomous driving', 'summary': 'Explains the use of supervised learning for autonomous driving using an artificial neural network, with training data collected from human drivers, and highlights the importance of strategic decision-making in machine learning projects.', 'duration': 473.187, 'highlights': ['The chapter demonstrates the use of supervised learning with an artificial neural network for autonomous driving, where training data is collected from human drivers, and the algorithm learns to mimic the human-selected driving direction.', 'Dean Pomelo showcases the significance of strategic decision-making in machine learning projects, emphasizing the crucial decisions related to data collection, choice of learning algorithms, and architectural choices for neural networks.', "The video illustrates the training process using backpropagation learning algorithm, showing how the neural network's outputs become less of a white smear and start to mimic the human-selected driving direction as it learns.", 'The chapter also discusses the effectiveness of different machine learning teams in applying learning algorithms, highlighting the importance of strategic decision-making and the potential impact on project outcomes.']}], 'duration': 645.119, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I2975999.jpg', 'highlights': ['The support vector machine algorithm can handle an infinite number of input features, providing a relatively effective learning algorithm for some problems.', 'Supervised learning involves training with inputs x and labels y to find a mapping for new x to the most appropriate output y, forming the basis of learning algorithms.', 'It is challenging to plot high-dimensional data with many features, making it difficult to visualize and analyze the data.', 'The chapter demonstrates the use of supervised learning with an artificial neural network for autonomous driving, where training data is collected from human drivers, and the algorithm learns to mimic the human-selected driving direction.', 'Dean Pomelo showcases the significance of strategic decision-making in machine learning projects, emphasizing the crucial decisions related to data collection, choice of learning algorithms, and architectural choices for neural networks.', 'The chapter also discusses the effectiveness of different machine learning teams in applying learning algorithms, highlighting the importance of strategic decision-making and the potential impact on project outcomes.', "The video illustrates the training process using backpropagation learning algorithm, showing how the neural network's outputs become less of a white smear and start to mimic the human-selected driving direction as it learns."]}, {'end': 4081.962, 'segs': [{'end': 3663.006, 'src': 'embed', 'start': 3621.118, 'weight': 4, 'content': [{'end': 3628.961, 'text': 'we want to help you become more systematic in driving machine learning as a- as a systematic engineering discipline.', 'start': 3621.118, 'duration': 7.843}, {'end': 3634.643, 'text': "So that when one day you're working on machine learning project, you can efficiently figure out what to do next, right?", 'start': 3629.021, 'duration': 5.622}, {'end': 3641.225, 'text': 'Um, and I sometimes make an analogy to how um to uh, uh to- to- to software engineering.', 'start': 3635.103, 'duration': 6.122}, {'end': 3648.511, 'text': 'Um, you know, I- many years ago, I had a friend, um, that would debug code by compiling it.', 'start': 3641.845, 'duration': 6.666}, {'end': 3655.438, 'text': 'And then, um, uh, this friend would look all these syntax errors, right? That, you know, C++ compiler outputs.', 'start': 3649.271, 'duration': 6.167}, {'end': 3661.885, 'text': 'And they thought that the best way to eliminate the errors is to delete all the lines of code with syntax errors.', 'start': 3655.959, 'duration': 5.926}, {'end': 3663.006, 'text': 'And that was their first seriously.', 'start': 3661.945, 'duration': 1.061}], 'summary': 'Transform machine learning into systematic engineering, drawing analogy to software engineering for efficient project execution.', 'duration': 41.888, 'max_score': 3621.118, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3621118.jpg'}, {'end': 3746.93, 'src': 'embed', 'start': 3704.313, 'weight': 1, 'content': [{'end': 3706.854, 'text': "And then, and then because he's so experienced it works.", 'start': 3704.313, 'duration': 2.541}, {'end': 3713.856, 'text': "But I think, um, what we're trying to do with the discipline of machine learning is to evolve it from a black magic, tribal knowledge,", 'start': 3706.914, 'duration': 6.942}, {'end': 3718.238, 'text': 'experience-based thing to a systematic engineering process, right?', 'start': 3713.856, 'duration': 4.382}, {'end': 3723.98, 'text': 'And so, um later this quarter, as we talk about machine learning strategy or talk about learning theory,', 'start': 3718.318, 'duration': 5.662}, {'end': 3729.342, 'text': 'we try to systematically give you tools on how to, um uh, go about strategizing.', 'start': 3723.98, 'duration': 5.362}, {'end': 3737.546, 'text': 'uh, so you can be very efficient in um how you- how you yourself or how you can lead a team to build an effective learning system.', 'start': 3729.842, 'duration': 7.704}, {'end': 3746.93, 'text': "Because I- I don't want you to be one of those people that you know waste six months on some direction that maybe could have relatively quickly figured out what's not promising.", 'start': 3737.566, 'duration': 9.364}], 'summary': 'Evolve machine learning from experience-based to systematic process, providing tools for efficient strategizing.', 'duration': 42.617, 'max_score': 3704.313, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3704313.jpg'}, {'end': 3845.636, 'src': 'heatmap', 'start': 3747.03, 'weight': 0.718, 'content': [{'end': 3755.013, 'text': "Or maybe one last analogy, if you, um, if you're used to optimizing code, right, making code run faster, Not sure how many of you have done that.", 'start': 3747.03, 'duration': 7.983}, {'end': 3762.415, 'text': 'Uh, uh, uh, less experienced software engineers will just dive in and optimize the code to try to make it run faster, right?', 'start': 3755.673, 'duration': 6.742}, {'end': 3765.216, 'text': "Let's take the C++ and code an assembly or something.", 'start': 3762.435, 'duration': 2.781}, {'end': 3772.398, 'text': 'But more experienced people will run a profiler to try to figure out what part of your code is actually a bottleneck and then just focus on changing on that.', 'start': 3765.636, 'duration': 6.762}, {'end': 3780.925, 'text': 'So, uh, one of the things hope to do this quarter is uh, uh, uh, convey to you some of these more systematic engineering principles, right?', 'start': 3773.118, 'duration': 7.807}, {'end': 3781.846, 'text': 'And yeah.', 'start': 3780.965, 'duration': 0.881}, {'end': 3787.911, 'text': "Oh, and then actually this is, uh, uh, uh, yeah, actually I've been, um, I've been, I've been writing.", 'start': 3781.946, 'duration': 5.965}, {'end': 3793.396, 'text': 'So actually, so, uh, how many of you have heard of machine learning yearning? Oh, just a few of you.', 'start': 3787.931, 'duration': 5.465}, {'end': 3800.642, 'text': 'Interesting. So, actually to to, if any of you are interested, um, just in my uh spare time.', 'start': 3793.556, 'duration': 7.086}, {'end': 3809.237, 'text': "Uh, I've been writing a book, um, uh, to try to codify systematic engineering principles for machine learning.", 'start': 3802.276, 'duration': 6.961}, {'end': 3816.658, 'text': 'And so if you, uh, uh, and, and so, uh, if you want a, you know, free draft copy of the book, sign up for a mailing list here.', 'start': 3809.317, 'duration': 7.341}, {'end': 3819.259, 'text': 'I tend to just write stuff and put it on the Internet for free, right?', 'start': 3816.798, 'duration': 2.461}, {'end': 3825.12, 'text': 'So if you want a free draft copy of the book, uh, uh, uh, you know, go to this website.', 'start': 3819.279, 'duration': 5.841}, {'end': 3829.121, 'text': 'uh, enter your email address and the website will send you a copy of the book.', 'start': 3825.12, 'duration': 4.001}, {'end': 3831.241, 'text': "They'll talk a little bit about these engineering principles as well.", 'start': 3829.181, 'duration': 2.06}, {'end': 3834.066, 'text': 'All right.', 'start': 3833.786, 'duration': 0.28}, {'end': 3836.869, 'text': 'So, uh, so first subject, machine learning.', 'start': 3834.367, 'duration': 2.502}, {'end': 3838.61, 'text': 'Second subject, learning theory.', 'start': 3837.029, 'duration': 1.581}, {'end': 3845.636, 'text': "Um and uh, the third major subject we'll talk about is, uh, deep learning, right?", 'start': 3839.471, 'duration': 6.165}], 'summary': 'Experienced engineers optimize code by identifying bottlenecks. author offers a free draft of a book on systematic engineering principles for machine learning.', 'duration': 98.606, 'max_score': 3747.03, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3747030.jpg'}, {'end': 3916.271, 'src': 'embed', 'start': 3883.223, 'weight': 2, 'content': [{'end': 3888.546, 'text': 'So, uh, other than deep learning, slash- after- after deep learning, slash, neuro- neural networks.', 'start': 3883.223, 'duration': 5.323}, {'end': 3895.11, 'text': "the other- the- the four- fourth of the five major topics we'll cover will be um, unsupervised learning.", 'start': 3888.546, 'duration': 6.564}, {'end': 3908.417, 'text': 'Um, so what is unsupervised learning? So you saw me draw a picture.', 'start': 3895.71, 'duration': 12.707}, {'end': 3916.271, 'text': 'like this just now, right? And this would be a classification problem like the, uh, tumor, malignant, benign problem.', 'start': 3910.666, 'duration': 5.605}], 'summary': 'Covered deep learning, neural networks, and unsupervised learning in the course.', 'duration': 33.048, 'max_score': 3883.223, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3883223.jpg'}, {'end': 3968.369, 'src': 'embed', 'start': 3940.365, 'weight': 0, 'content': [{'end': 3946.807, 'text': 'Um, and so in this dataset it looks like there are two clusters and an unsupervised learning algorithm which you learn about,', 'start': 3940.365, 'duration': 6.442}, {'end': 3952.168, 'text': 'called k-means clustering will discover this, um, this structure, in the data.', 'start': 3946.807, 'duration': 5.361}, {'end': 3958.62, 'text': 'Um, other examples of personalized learning, you know, if, if you, actually Google News is a, is a very interesting website.', 'start': 3953.015, 'duration': 5.605}, {'end': 3961.262, 'text': 'Sometimes I use it to look up, right, latest news.', 'start': 3958.64, 'duration': 2.622}, {'end': 3962.483, 'text': 'Uh, this is an old example.', 'start': 3961.483, 'duration': 1}, {'end': 3968.369, 'text': "But Google News every day, uh, crawls or reads, uh, uh, uh, I don't know.", 'start': 3962.904, 'duration': 5.465}], 'summary': 'K-means clustering identifies two clusters in the dataset, and google news is an example of personalized learning.', 'duration': 28.004, 'max_score': 3940.365, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3940365.jpg'}], 'start': 3621.118, 'title': 'Applying systematic engineering principles in machine learning', 'summary': 'Stresses the significance of a systematic approach in machine learning, drawing parallels to software engineering. it discusses debugging code and learning algorithms, and highlights the practical applications of unsupervised learning in different domains.', 'chapters': [{'end': 3663.006, 'start': 3621.118, 'title': 'Systematic approach to machine learning', 'summary': 'Emphasizes the importance of applying a systematic approach to machine learning projects, drawing an analogy to software engineering and highlighting the significance of efficient problem-solving techniques, as illustrated by the example of debugging code.', 'duration': 41.888, 'highlights': ['The importance of becoming more systematic in driving machine learning as a systematic engineering discipline to efficiently figure out the next steps for machine learning projects.', 'Emphasizing the analogy between systematic approach in machine learning and software engineering for efficient problem-solving techniques.', 'Illustrating the significance of efficient problem-solving techniques through an example of debugging code, where a friend attempted to eliminate syntax errors by deleting the lines of code with errors.']}, {'end': 4081.962, 'start': 3663.326, 'title': 'Engineering principles in machine learning', 'summary': 'Emphasizes the transition from black magic to systematic engineering in machine learning, introducing the importance of debugging learning algorithms and the systematic engineering principles for efficient learning systems. it also covers the major subjects of machine learning, learning theory, deep learning, and unsupervised learning, highlighting the practical applications of unsupervised learning in various domains.', 'duration': 418.636, 'highlights': ['The transition from black magic to systematic engineering in machine learning is emphasized, focusing on the significance of debugging learning algorithms and the systematic engineering principles for efficient learning systems.', 'The practical applications of unsupervised learning in various domains are highlighted, including clustering algorithms for organizing computing clusters, identifying cohesive communities within social networks, and market segmentation for targeted marketing.', 'The major subjects of machine learning, learning theory, deep learning, and unsupervised learning are covered, with a specific focus on deep learning and its rapid advancements, as well as the concept and applications of unsupervised learning in identifying interesting structures in data with no labels.', 'The importance of efficiently strategizing and building effective learning systems is emphasized, aiming to prevent wasting time on unpromising directions and to convey more systematic engineering principles to optimize the process.']}], 'duration': 460.844, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I3621118.jpg', 'highlights': ['The practical applications of unsupervised learning in various domains are highlighted, including clustering algorithms for organizing computing clusters, identifying cohesive communities within social networks, and market segmentation for targeted marketing.', 'The importance of efficiently strategizing and building effective learning systems is emphasized, aiming to prevent wasting time on unpromising directions and to convey more systematic engineering principles to optimize the process.', 'The major subjects of machine learning, learning theory, deep learning, and unsupervised learning are covered, with a specific focus on deep learning and its rapid advancements, as well as the concept and applications of unsupervised learning in identifying interesting structures in data with no labels.', 'The transition from black magic to systematic engineering in machine learning is emphasized, focusing on the significance of debugging learning algorithms and the systematic engineering principles for efficient learning systems.', 'The importance of becoming more systematic in driving machine learning as a systematic engineering discipline to efficiently figure out the next steps for machine learning projects.', 'Emphasizing the analogy between systematic approach in machine learning and software engineering for efficient problem-solving techniques.', 'Illustrating the significance of efficient problem-solving techniques through an example of debugging code, where a friend attempted to eliminate syntax errors by deleting the lines of code with errors.']}, {'end': 4514.226, 'segs': [{'end': 4112.191, 'src': 'embed', 'start': 4082.742, 'weight': 3, 'content': [{'end': 4090.164, 'text': 'Uh, and and, and actually many years ago my friend Andrew Moore, uh, uh, was using this type of data for astronomical data analysis,', 'start': 4082.742, 'duration': 7.422}, {'end': 4092.145, 'text': 'to group together galaxies.', 'start': 4090.164, 'duration': 1.981}, {'end': 4099.807, 'text': "Question? Oh, is all unsupervised learning clustering? No, it's not.", 'start': 4092.605, 'duration': 7.202}, {'end': 4104.209, 'text': 'So unsupervised learning broadly is the concept of using unlabeled data.', 'start': 4099.987, 'duration': 4.222}, {'end': 4112.191, 'text': "So just X and finding interesting things about it, right? Um, so, uh, for example, uh, actually here's- shoot.", 'start': 4104.349, 'duration': 7.842}], 'summary': 'Unsupervised learning used for astronomical data analysis to group galaxies.', 'duration': 29.449, 'max_score': 4082.742, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I4082742.jpg'}, {'end': 4160.283, 'src': 'embed', 'start': 4130.609, 'weight': 2, 'content': [{'end': 4131.649, 'text': 'Let me think how to explain this.', 'start': 4130.609, 'duration': 1.04}, {'end': 4133.89, 'text': 'Uh, you know, cocktail party problems.', 'start': 4132.21, 'duration': 1.68}, {'end': 4137.591, 'text': "I'll, I'll try to do the demo when we can get audio work on this laptop.", 'start': 4134.011, 'duration': 3.58}, {'end': 4146.236, 'text': "There's a problem where, um uh, if you have a noisy room and you stick multiple microphones in the room and record overlapping voices, uh,", 'start': 4137.971, 'duration': 8.265}, {'end': 4147.096, 'text': 'so there are no labels.', 'start': 4146.236, 'duration': 0.86}, {'end': 4150.518, 'text': 'where you just have multiple microphones, an array of microphones.', 'start': 4147.096, 'duration': 3.422}, {'end': 4155.6, 'text': "in a room with lots of people talking, uh, how can you have the algorithm separate out the people's voices?", 'start': 4150.518, 'duration': 5.082}, {'end': 4160.283, 'text': "So that's an unsupervised learning problem, because, um, there are no labels.", 'start': 4156.22, 'duration': 4.063}], 'summary': 'Explaining unsupervised learning for separating voices in a noisy room with multiple microphones.', 'duration': 29.674, 'max_score': 4130.609, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I4130609.jpg'}, {'end': 4194.927, 'src': 'embed', 'start': 4169.535, 'weight': 5, 'content': [{'end': 4175.158, 'text': 'And one of the programming exercises you do, uh, later is if we have, you know, five people talking.', 'start': 4169.535, 'duration': 5.623}, {'end': 4183.062, 'text': "So each microphone records five people's overlapping voices, right? Because, you know, each microphone hears five people at the same time.", 'start': 4176.058, 'duration': 7.004}, {'end': 4185.182, 'text': 'How can you have an algorithm?', 'start': 4183.801, 'duration': 1.381}, {'end': 4189.084, 'text': 'separate out these voices so you get clean recordings of just one voice at a time?', 'start': 4185.182, 'duration': 3.902}, {'end': 4191.404, 'text': "So that's called the cocktail party problem.", 'start': 4189.604, 'duration': 1.8}, {'end': 4194.927, 'text': 'And the algorithm you use to do this is called ICO, Independent Components Analysis.', 'start': 4191.524, 'duration': 3.403}], 'summary': 'Algorithm ico solves cocktail party problem by separating overlapping voices.', 'duration': 25.392, 'max_score': 4169.535, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I4169535.jpg'}, {'end': 4260.999, 'src': 'embed', 'start': 4233.552, 'weight': 1, 'content': [{'end': 4237.735, 'text': 'It turns out you can learn analogies like that from unlabeled data, just from text on the Internet.', 'start': 4233.552, 'duration': 4.183}, {'end': 4239.537, 'text': "So that's also unsupervised learning.", 'start': 4237.756, 'duration': 1.781}, {'end': 4246.851, 'text': 'Okay Um, so after unsupervised learning, oh, and unsupervised learning.', 'start': 4240.418, 'duration': 6.433}, {'end': 4250.012, 'text': 'So, you know, machine learning is very useful today.', 'start': 4246.871, 'duration': 3.141}, {'end': 4256.736, 'text': 'Turns out that most of the recent wave of economic value created by machine learning is through supervised learning.', 'start': 4250.333, 'duration': 6.403}, {'end': 4260.999, 'text': 'Uh, but there are important use cases for unsupervised learning as well.', 'start': 4257.437, 'duration': 3.562}], 'summary': 'Unsupervised learning from unlabeled data on the internet, important for machine learning and creating economic value.', 'duration': 27.447, 'max_score': 4233.552, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I4233552.jpg'}, {'end': 4503.48, 'src': 'embed', 'start': 4458.213, 'weight': 0, 'content': [{'end': 4467.261, 'text': 'Um, I think that, uh, I, I, I think that, uh, uh, game playing has made for some remarkable stunts, a remarkable PR.', 'start': 4458.213, 'duration': 9.048}, {'end': 4471.144, 'text': "but I'm also, uh, equally excited, or maybe even more excited,", 'start': 4467.261, 'duration': 3.883}, {'end': 4475.407, 'text': 'about the inroads that reinforcement learning is making into robotics applications right?', 'start': 4471.144, 'duration': 4.263}, {'end': 4480.63, 'text': 'So I think, uh, Uh, I think, yeah, reinforcement has been proven to be fantastic for playing games.', 'start': 4475.427, 'duration': 5.203}, {'end': 4488.414, 'text': "It's also getting- making real traction in, um, optimizing robots and optimizing sort of, uh, the logistics system and things like that.", 'start': 4480.71, 'duration': 7.704}, {'end': 4492.875, 'text': 'So you learn about all these things.', 'start': 4490.894, 'duration': 1.981}, {'end': 4503.48, 'text': 'Um, last thing for today, uh, I hope that you will start to, uh, talk- meet people in the class, make friends, form project partners and study groups.', 'start': 4493.375, 'duration': 10.105}], 'summary': 'Reinforcement learning is making inroads in robotics applications, proving fantastic for playing games and optimizing logistics systems.', 'duration': 45.267, 'max_score': 4458.213, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I4458213.jpg'}], 'start': 4082.742, 'title': 'Unsupervised learning', 'summary': 'Covers the concept of unsupervised learning and its applications in astronomical data analysis, the cocktail party problem, analogies from unlabeled text data, optimizing robots and game playing, with reinforcement learning gaining traction in robotics and logistics applications.', 'chapters': [{'end': 4189.084, 'start': 4082.742, 'title': 'Unsupervised learning applications', 'summary': 'Covers the concept of unsupervised learning, including its application in astronomical data analysis and the cocktail party problem, which involves separating overlapping voices recorded by multiple microphones.', 'duration': 106.342, 'highlights': ['Unsupervised learning used for astronomical data analysis to group galaxies together.', 'Cocktail party problem: separating overlapping voices recorded by multiple microphones in a noisy room without labels.', 'Programming exercise involving the algorithm to separate overlapping voices of five people recorded by multiple microphones.']}, {'end': 4514.226, 'start': 4189.604, 'title': 'Unsupervised and reinforcement learning', 'summary': 'Covers unsupervised learning, reinforcing learning, and their applications, including analogies from unlabeled text data, optimizing robots and game playing, with reinforcement learning gaining traction in robotics and logistics applications.', 'duration': 324.622, 'highlights': ['Reinforcement learning is making real traction in optimizing robots and logistics systems.', "Learning analogies like 'a man is a woman, a king is a queen' from unlabeled data is an example of unsupervised learning.", 'Reinforcement learning has been proven to be fantastic for playing games, and it is also making real traction in optimizing robots and logistics systems.']}], 'duration': 431.484, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/jGwO_UgTS7I/pics/jGwO_UgTS7I4082742.jpg', 'highlights': ['Reinforcement learning is making real traction in optimizing robots and logistics systems.', "Learning analogies like 'a man is a woman, a king is a queen' from unlabeled data is an example of unsupervised learning.", 'Cocktail party problem: separating overlapping voices recorded by multiple microphones in a noisy room without labels.', 'Unsupervised learning used for astronomical data analysis to group galaxies together.', 'Reinforcement learning has been proven to be fantastic for playing games, and it is also making real traction in optimizing robots and logistics systems.', 'Programming exercise involving the algorithm to separate overlapping voices of five people recorded by multiple microphones.']}], 'highlights': ['The demand for AI and machine learning skills is vast, with the majority of students applying for such skills.', 'The rise of machine learning has led to numerous opportunities to apply learning algorithms in both industry and academia.', 'The speaker led the AI transformations of Google and Baidu, significantly impacting their status as prominent AI companies today.', 'The class has helped several generations of Stanford students become experts in machine learning, who have built products and services and startups.', 'The number of people doing valuable machine learning projects has significantly increased over the past few years.', 'The teaching team, including the class coordinator and co-head TAs, provide technical expertise and support to students, with a large TA team spanning various machine learning application areas.', 'The relevance of machine learning has expanded to non-tech companies, creating a significant need to apply machine learning tools.', "The course website showcases previous year's projects, demonstrating the application of machine learning to a wide range of areas, including diagnosing cancer, creating art, and engineering applications.", 'The course will be using Gradescope for online grading and updating syllabus and content.', 'The switch to Python and take-home midterm are significant changes for the course.', 'The support vector machine algorithm can handle an infinite number of input features, providing a relatively effective learning algorithm for some problems.', 'The practical applications of unsupervised learning in various domains are highlighted, including clustering algorithms for organizing computing clusters, identifying cohesive communities within social networks, and market segmentation for targeted marketing.', 'Reinforcement learning is making real traction in optimizing robots and logistics systems.', 'The chapter explains supervised learning through housing price prediction.', 'The class project is a key component, aiming to leave students well-qualified to undertake a meaningful machine learning project, supported by TAs.']}