Live data from Hacker News

Ask HN: What's the best lecture series you've seen?

news.ycombinator.com

51–60 of 287 posts

Re: Ask HN: What's the best lecture series you've seen?

#51
post #30
post #12

Andrej Karpathy's "Neural Networks: From Zero to Hero". https://karpathy.ai/zero-to-hero.html Just watch the first lecture and you won't be able to not watch the rest. It starts with making your own autograd engine in 100 lines of python, similar to PyTorch and then builds up to a GPT network. He's one of the best in the field, founder of OpenAI, then Director of AI at Tesla. Nothing like the scam tutorials that just…

[flagged]

Let me help you there. Have you considered writing: "I think this lecture series is overrated, because $SERIES_A and $SERIES_B is much better, because ..."

Or you could explain why you dislike the communication style, the pace or anything else that others could also relate to.

This way we can just guess, and guessing includes bad options like "dev_0 just had a petty conflict with the lecturer" or "dev_0 tried themselves to teach this and did worse so it is envy".

So if you want to provide value, please explain yourself or just don't comment with one word in the future. If you don't want to provide value, consider commenting on youtube instead.

Re: Ask HN: What's the best lecture series you've seen?

#53
post #32

Gilbert Strang's lectures on Linear Algebra [1]. Prof Strang is an amazing lecturer with a unassuming style. He's the expert on Linear Algebra and makes the topic so much approachable. Ken Joy's lectures on Computer Graphics [2]. Prof Joy is another amazing lecturer, making the Computer Graphics topics seeming easy. Stanford CS221 Learn AI (2019) by Percy Liang and Dorsa Sadigh. Both professors are great lecturers. A…

Can you articulate on the first one? In the last semester I tutored for a Linear algebra course which used his book and it was a nightmare. Ideas seems to be presented in reverse order and a lot of students ended up having trouble understanding basic concepts.

I haven't used his book, just watched the Youtude videos. His teaching went slowly with simple examples. He explained them really well.

There're a lot of materials, much more than one would care. He covered many topics. I just jumped to the ones I needed to learn at the time. I had Linear Algebra background so most were just a refresher for me. As a student attending the class the first time, it might be overwhelming to learn all those material in a semester.

Re: Ask HN: What's the best lecture series you've seen?

#54
Geoff Hinton's legendary Coursera course on neural nets. It came out around the same time as AlexNet winning the ImageNet competition which sparked the current deep learning revolution.

At the time it was cutting edge to the point where he introduced a previously undescribed optimization method (RMSProp) that was subsequently used in papers, citing the lecture slides as their reference! But still accessible to anyone with basic college math. Of course it doesn't have any of the new stuff like transformers or diffusion models, but I still consider it as giving a good foundation for understanding backprop and neural nets.

Unlike every other AI course at the time it didn't try to teach you about all the other types of machine learning. Neural nets only. After taking it I was able to apply neural nets at work with pretty great results. Also, it gave me one of my favorite quotes: "To deal with hyper-planes in a 14-dimensional space, visualize a 3-D space and say 'fourteen' to yourself very loudly. Everyone does it."

Re: Ask HN: What's the best lecture series you've seen?

#55

Norman Wildberger's "Wild Linear Algebra" series https://www.youtube.com/playlist?list=PLIljB45xT85BhzJ-oWNug... His geometry-centric approach to linear algebra was exactly what I needed to finally grok the subject. Topics like matrix multiplication and discriminants went from "why are they defined like this? it makes no sense?" to "of course that's how you multiply matrices because it's the only logical answer". It'…

It's not just imaginary numbers, the ultrafinitist rabbit hole runs deep. But there's nothing inherently wrong with it, it's just a different set of axioms than everyone else uses. With those constraints you can still do meaningful things by other means (checkout his rational trigonometry stuff), at the expense of more cumbersomeness.

Re: Ask HN: What's the best lecture series you've seen?

#56
Not even a second of thought: Sussman's & Abelson's SICP https://www.youtube.com/watch?v=-J_xL4IGhJA&list=PLE18841CAB...

especially the section on streams (https://www.youtube.com/watch?v=JkGKLILLy0I&list=PLE18841CAB...)

It changed my outook on programming by pi degrees and I feel it is more needed then ever.

Re: Ask HN: What's the best lecture series you've seen?

#57

Intro to Database Systems by Andy Pavlo - https://www.youtube.com/watch?v=oeYBdghaIjc&list=PLSE8ODhjZX... MIT 6.824 Distributed Systems by Robert Morris - https://www.youtube.com/watch?v=cQP8WApzIQQ&list=PLrw6a1wE39...

I love the Robert Morris’ lectures. Best ever in DS.

Re: Ask HN: What's the best lecture series you've seen?

#60
post #48

Timothy Snyder: The Making of Modern Ukraine. [0] gives you invaluable understanding about much more than just Ukraine. MIT 16.885J Aircraft Systems Engineering, Fall 2005 [1] - the aircraft they focus on is the Space Shuttle. Amazingly demystifying. Some of the actual early designers talk there. [0] - https://www.youtube.com/watch?v=bJczLlwp-d8&list=PLh9mgdi4rN... [1] - https://www.youtube.com/watch?v=iiYhQtGpRhc&li…

I second to this [0]. This was basically European history course from times of Vikings till present day.
Post reply on HN