This is so great for making these valuable ressources completely free to us. Would you consider uploading courses for civil engineering department? Thank you 🙏🏽
it's not always that you want an unlimited dataset in machine learning. Sometimes you want to limit your dataset. For example if you're trying to figure out if you're in a stock market recession or expansion you don't want to go over 2 years (+- months) of data because you'll end up in another cycle. Symmetry and Domain limits are important in machine learning. My point is sometimes too much data is a waste of time and resources
Mostly overview, not much nitty-gritty. 16:06 The primacy of data sets. 37:03 Clearly demonstrated by that famous 1983 documentary WarGames. 42:55 Machine Learning and Neural Networks.
@@shivang581 The prerequisites listed in the syllabus are: 18.06 Linear Algebra and familiarity with MATLAB®. See the course on MIT OpenCourseWare for more info at: ocw.mit.edu/RES-LL-005IAP20. Best wishes on your studies!
Great video guys and handfull overview of AI and ML, well summarize and explanations. clear slides and precise examples. each question people did cleared doubts after Vijay's answers. Thank you MIT team Mr Jeremy and Vijay.
artificial neural networks, complexity-emergence, order, information theory, qbit, machine vision... all are on the way... but forget all. just learn python;)
@@MA-nv6kc It's easy, but if you want to learn it fast, buy a Udemy course or try to find free courses so you can progress faster. Obviously the hardest part of python is the understanding of the math you are going to use, if you have high level math skills, python will be "easy" for you, but still you need some course to move/learn faster
Thanks a lot for describing such a big comprehensive AI information in such a way that I can easily see and understand the forest of AI, especially machine learning !!!
Autonomous vehicles are pretty safe. While this is debatable and certainly a matter of opinion, I (and many others) would argue that the barrier to wider adoption isn't due to the navigation systems being unsafe; it's our unwillingness to accept anything short of perfection. It's not really a fair comparison (Human vs. AI drivers) if we only count strikes against one side.
There's this book called super intelligence by Nick Bostrom thats mainly about the dangers and paths of A.I. Elon Musk recommends it and I've read some of it too, its pretty interesting.
Great introductory video! The presentation covers just about everything there is to know about Artificial Intelligence and Machine Learning up until the present time. I wonder if the general public can download his slides somewhere? I'd love to have a detailed look at all those slides.
Thank you MIT for allowing us all to enjoy your quality education at a fraction of the cost :) Your content has been very helpful to me during Covid lockdown when I have to study at home.
thanks a lot MIT. If it is possible,could you make the videos of 18.04 Complex Variable with applications?I really want to watch that course.Thank you very much.😁
The prerequisites listed in the syllabus: 18.06 Linear Algebra and familiarity with MATLAB®. See the resource on MIT OpenCourseWare for more info at: ocw.mit.edu/RES-LL-005IAP20. Best wishes on your studies!
Watch it again but the next time Reeelaaaxxxxx. Just listen and hit Pause if you need to look at a slide closer or run the click back on the timeline in the video and repeat anything he says you aren't quite getting. If it sill stumps you, get a piece of paper and a pen and go to the spot in his lecture you started getting confused. Listen for keywords you don't know, write them down, and run a search on them. You may find you make an indirect 'leap' in your understanding merely by comprehending those words *in the context of their use*. Or, you can write diem what your problem is. "When he starts talking about X, I get confused. It's X I don't understand OR I don't understand X in the context he's using it. Is that what's confusing me?". I guess I'm saying that if you can identify when and where you Begin to get confused, you may find your problem isn't with the topic or the lexicon, but that your attempt to understand came Before your brain was prepared To understand. IOW, if you don't speak Greek it won't matter WHAT the person is trying to tell you, because you literally don't know the meanings of the words (labels). So a word like Cat "points to" a place in your brain that has ALL the features of Cats you've learned about and have experience with: 4 legs, long tail, average adult cat size, whiskers, vertical pupils, triangle ears, fur(soft), meows, etc etc. But if you hear a word or phrase you've NEVER heard, it could be a new wordand you could be an expert in your field but your don't live in the time zone where big science stuff breaks out in time right so there you are,?an expert, a new word in your professional lexicon but you don't know it! So your peer uses it, you get confused, they explain it to you, you're embarassed. ALL of THAT is highly complex behaviors/results and systems at work, at blisteringly fast speeds. The very process of learning itself, which is what a lecture like this tells me, is not a static series of methodical, linear actions is occurring in learning but that there's all sorts of pre and post events tied to a minor event and you have millions of these minor events not occurring in any typically plodding way of "first This, then That" but rather more burst_events like millions of Roman candles, and each burst is made up of several trailing rays moving away from the burst but in human learning it would be more like each burst events rays that come out from it were sidewinding and crossing the paths of rays trajectoried from their burst event. So I say Cat and without anyone being able to tell, your mind searches for Cat to make an association and a check that your brain can and does know what a Cat is. Cool, you can understand my language. But in learning Experience To Concept, Experience To Object, and Experience to Topic (I'm arbitrariy making up phrases here btw, and I've capitalized them so you can tell there's some special text emphasis going on, and now you know what it is ). So with Cat, let's say: 1. You have experience To the Concept of Cat (non-human animal) 2. You have experience to the Object of Cat (physical features) 3. You have experience to the topic of Cat, (maybe you had a kitten as a kid and how you feel about that experience might impact how effectively you continue to learn about Cats, maybe you have superstitions about Cats, etc etc). The man giving this lecture is speaking English. Do you have any issues at all with English? Maybe it's not your first language? Or is it your first language but you don't know anything at all about Computer Science other than it is a field of study and research? If that's the case, then did you listen to his entire lecture but you understood literally nothing he said or...? Do you know that it's normal for a speaker to be an expert on the topic and the people listening are not? And in that case, do you think its possible at least a few in the audience probably missed things he said or were confused by them, but group dynamics are such they'll not admit to it and improve Learning but rather tend to stay silent, and then forget to address themselves to what they did NOT understand? It's quite frequent for people to lie to each other about what they don't understand. You say it went over your head. Its nearly a 2 hour lecture. So are your saying you sat paying close attention for 2 hours to this man and it ALL went over your head? You have no clearer understanding about differences between AI learning and Machine learning? After the slides and his lecture, it all "went over your head"? Sir or Madam, at this point I want you to cut yourself some slack. For one, it's very difficult being a human being. For two, communicating with each other, even in the most ideal circumstances, is THE hardest thing we do together. Finally, I hope in this comment you will be able to asses yourself more fairly in the future. Maybe you're too hard on yourself. Or maybe you're too easy and make too many excuses, too frequently, for under performing and it feeds into a complex self-sabotage crisis that feeds into anxiety that feeds into system overload that feeds into failure that feeds into self-confidence and this is your 'endless loop' bug. The feature of your brain that is still curious compels you to watch this video, right? But the bug is corrupting the signal trying to stay live long enough to steer you To Learning. I'm saying that SOME part of you WANTED to understand. Why? Do you know? And, are you aware this comment has used certain aspects of the videos topic when replying to you? 😁 Computer Science is a VERY big umbrella with lots of cool things to study and participate in. Maybe you should cut yourself some slack and go ahead and start studying it? Or maybe you're more the Computer Engineering type? Well pick one and them go to MITs website and look up what their CompSci classes are and what textbooks then try to get the books and away you go. No muss, no fuss, no college debt. If you don't *need* the degree, NEVER buy any universities debt! You are allowed to study & learn apart from Universities. You are allowed to study & learn apart from degrees. As a child, I had read and learned enough that I decided I wanted to become an autodidactic dillatante, and that's precisely what I achieved by 21. Best thing I ever put myself through. Philosophy, History(World/Military/Americas as well as an immense number of "History Of...Specific Topic"), Maths (i stopped teaching myself at Calculus I'm afraid), Chemistry, Biology, Marine biology, French, Latin, Spanish, German, Anatomy & Physiology, sevetal extensible and non-extensible computer languages, system architectures, large object models, and on and on it goes my friend. Your study, you learn, you put it into practice for awhile by Trying and Failing a lot then comes your Eureka moment. It all makes sense to you. For me, it's that I'm alive to learn. When I tire of learning I know I'll be near my check out time for Hotel Earth. And all I learned will be lost and I won't be sad about it for even a second because all the confidence and contentment that comes from Knowing pales in comparison to the excitement and mystery and surprise, and yes the aggravation and tedium and suffering, if you will, is there as well, they are part of the price of the ticket after all, being immersed in and wholly consumed in the act of Learning. If you can catch what I'm saying to you, then you may realize I've paid pretty close attention to Learning :) And *this* particular path is the reason I know as much as I do about so many different topics or why it's difficult for religous fanatics to be around me (and vice versa), but this isnt me humble bragging so much as poking you with a long stick to say,"Why would you let anything go over your head you obviously had to have had SOME interest in catching." ?
Hi, No offence. For activation function, is it some kind of luck to try out a good one? I mean it seems like trying out different functions and different params, then, ok, a better one has been found. Isn't it?
