Best Machine learning & deep learning Videos
「Machine learning & deep learning」下面有131部被不同的人、在不同月份推荐过的视频。挑的时候先看对象:入门25部,适合还没学过的人;进阶51部,适合已经懂术语的人;普通向38部,按兴趣就能看;更硬的17部,留给想看推导或细节的人。这一类数量最多的是演讲,所以它们的共同点是一位讲者把一个判断讲完,例子是为了撑住这个判断。常出现的频道包括Andrej Karpathy、3Blue1Brown、Google TechTalks。目录给这个主题划的范围是:Machine learning: neural networks, deep learning, backpropagation, training, optimization, reinforcement learning, AlphaGo, ML research lectures, Karpathy, Hinton, LeCun, Stanford CS231n.如果只看三部,就从《The spelled-out intro to neural networks and backpropagation: building micrograd》开始,它负责把问题讲出画面;然后看《But what is a neural network? | Deep learning chapter 1》,换一种讲法或例子;最后看《AlphaGo - The Movie | Full award-winning documentary》,留给已经有基础的人。页面按每页24部切开,顺序看的是被推荐跨越了多少个月,不是播放量。每条标题下面那一句是这部片子自己的内容说明,用来区分它在讲定义、过程、演示还是观点。点进去能看到推荐原话、原文链接和适合谁看。看的时候把强烈推荐当成有人明确叫你去看,把引用当成只是拿来举例,把批评留下来对照,不要把每条评论都当成同一力度。

Peter Norvig - The Unreasonable Effectiveness of Data
Peter Norvig's UBC distinguished lecture on how massive amounts of simple data, such as for language modeling, can yield understanding that hand-built models could not.

Lecture 1: Introduction to Information Theory
The first lecture of David MacKay's Cambridge course on information theory, pattern recognition and neural networks.

Infinite Mario AI - Long Level
A bot built by Robin Baumgarten plays Infinite Mario using A* path-finding, shown as a red planned path that updates as new obstacles appear.

RAAIS 2017 - Simon Knowles, Co-Founder and CTO of Graphcore
Simon Knowles, co-founder and CTO of Graphcore, talks about "intelligence processors", chips designed specifically for machine learning.

CS480/680 Lecture 19: Attention and Transformer Networks
A university lecture from CS480/680 by Pascal Poupart on attention mechanisms and transformer networks, worked through on a chalkboard.

The spelled-out intro to language modeling: building makemore
Andrej Karpathy builds a bigram character-level language model from scratch, introducing torch.Tensor and the framework of training, sampling and loss evaluation.

Tesla Autonomy Day
Tesla's roughly four-hour Autonomy Day investor event covers its in-house self-driving chip, vision-only approach, fleet data engine and robotaxi plans, with talks by Andrej Karpathy and others.

Stanford Seminar: Beyond Floating Point: Next Generation Computer Arithmetic
John Gustafson presents posits, a proposed drop-in replacement for IEEE 754 floats that is simpler in hardware and more flexible in precision.

Joscha: From Computation to Consciousness (31c3)
Joscha Bach's 31c3 talk argues that the mind should be understood as an information-processing system and that even the physical universe can be seen as computational.

Large Language Models from scratch
A short cartoon explaining how language models like ChatGPT work, from autocomplete and conditional probabilities to neural networks, gradient descent and backpropagation.

AI Language Models & Transformers - Computerphile
Rob Miles explains on Computerphile how language models generate plausible text and how transformers improved on recurrent networks, with a look at what comes next.

Jeff Hawkins: Thousand Brains Theory of Intelligence | Lex Fridman Podcast #25
Lex Fridman interviews Jeff Hawkins, founder of Numenta and author of On Intelligence, about the Thousand Brains Theory and reverse-engineering the neocortex.

Practical Deep Learning for Coders: Lesson 1
Lesson 1 of fast.ai's Practical Deep Learning for Coders, taught by Jeremy Howard.

Sacha Arnoud, Director of Engineering, Waymo - MIT Self-Driving Cars
Sacha Arnoud, Waymo's Director of Engineering, gives a lecture in MIT's 2018 deep learning for self-driving cars course on the rise of machine learning in autonomous vehicles.

Tesla AI Day 2022
Tesla's full AI Day 2022 presentation, covering the Tesla Bot hardware and software, Full Self-Driving planning and perception, the occupancy network, data labeling, simulation and Dojo training infrastructure.

AlphaFold - The Most Useful Thing AI Has Ever Done
A Veritasium video on AlphaFold, DeepMind's protein structure prediction system, and the protein design work it enabled.

Alpha Zero's "Immortal Zugzwang Game" against Stockfish
agadmator walks through a game in which DeepMind's AlphaZero beat Stockfish, ending in a famous zugzwang that leaves the engine helpless.

HC2023-K2: Hardware for Deep Learning
Bill Dally of NVIDIA gives a Hot Chips 2023 keynote on the challenges of building hardware for deep learning.

Andrew Ng: Artificial Intelligence is the New Electricity
Andrew Ng speaks at the Stanford MSx Future Forum about how AI is transforming industry after industry, comparing its impact to electricity.

AI can't cross this line and we don't know why.
A Welch Labs explainer on neural scaling laws and the compute-efficient frontier, asking why model loss seems to hit a line it cannot cross.

Karl Sims - Evolved Virtual Creatures, Evolution Simulation, 1994
Karl Sims's 1994 demonstration of evolved virtual creatures: block-bodied creatures whose bodies and controllers are evolved through selection, mutation and recombination to swim, walk and compete.

Peter Norvig at Startup School 08
Peter Norvig, then Director of Research at Google, talks at Startup School 2008 about leveraging large amounts of data and about technical challenges in Google projects such as image search.

OpenAI Plays Hide and Seek…and Breaks The Game! 🤖
Two Minute Papers covers OpenAI's multi-agent hide-and-seek experiment, in which agents learned emergent tool use and found ways to break the game's rules.

Enhancing Photorealism Enhancement
A video presenting an Intel research paper on enhancing the photorealism of synthetic images.