/explore
|
You are here 
|
blog.ml.cmu.edu | ||
| | | | |
www.jeremykun.com
|
|
| | | | | In our last primer we saw the Fourier series, which flushed out the notion that a periodic function can be represented as an infinite series of sines and cosines. While this is fine and dandy, and quite a powerful tool, it does not suffice for the real world. In the real world, very little is truly periodic, especially since human measurements can only record a finite period of time. Even things we wish to explore on this blog are hardly periodic (for instance, image analysis). | |
| | | | |
windowsontheory.org
|
|
| | | | | Previous post: ML theory with bad drawings Next post: What do neural networks learn and when do they learn it, see also all seminar posts and course webpage. Lecture video (starts in slide 2 since I hit record button 30 seconds too late - sorry!) - slides (pdf) - slides (Powerpoint with ink and animation)... | |
| | | | |
francisbach.com
|
|
| | | | | [AI summary] This text discusses the scaling laws of optimization in machine learning, focusing on asymptotic expansions for both strongly convex and non-strongly convex cases. It covers the derivation of performance bounds using techniques like Laplace's method and the behavior of random minimizers. The text also explains the 'weird' behavior observed in certain plots, where non-strongly convex bounds become tight under specific conditions. The analysis connects theoretical results to practical considerations in optimization algorithms. | |
| | | | |
polukhin.tech
|
|
| | | A robot sitting next to a human in an office, trending on artstation, beautiful coloring, 4k, vibrant, blue and yellow, by DreamStudio | ||
