Deepmind artificial intelligence @ FDOT14
Channel:Elizr · Length:4:40 · Published:2014-04-19 · Views:288065
Summary
A short smartphone-recorded clip from a 2014 event in Paris where DeepMind CEO Demis Hassabis presents an AI learning to play old Atari games. The system learns from only the video signal and the score. Audio quality is poor.
Why HN recommends it
HN commenters use it as the canonical demo of deep reinforcement learning, noting that watching it teaches more than an article does.
you learn more about the tech from a youtube video than from an article — HN, 2014
Machine learning & deep learning · Game programming & engines
Subtitles (excerpt)
rather hundred gates so you can see it's pretty it's not very good but you can see it's sort of understanding of slowly you can see it's kind of understanding what it's supposed to be doing trying to move towards the ball so after an hour of training about 200 games so now you can see it's quantitatively better it's getting them all back around thirty or forty percent the time it's still not perfect but it's getting basically better now after two outs or 300 games it's better than any human can play this game so again this is the ball even when the ball comes back at very fast verticals so then we thought okay that's pretty good but let's just leave it training for another two hours see what happens and then this amazing thing happened it's all we're going to have lots of especially bigger titles and then send all round around somewhere not researchers learning amazing guys that working on this saw that that actually shot them because we did we haven't added in the sorts of capabilities like long-term memory at that point was that we thought was going to be needed to solve that and kind of get in that way and just to show you his general general capability so just wanna sort of one minute they're here of you know sort of dozens of other
On Hacker News
The Deepmind Atari Player is only 22KB of source code: [LINK] If you haven't seen the video it's remarkable: [VIDEO] — paulsutter, 2016-03
In psychology, this is called "Reinforcement"[1] and that goes back to Skinner (1938) and Thorndike (1911). In AI research its called "Reinforcement learning"[2]. One of the earliest papers relating reinforcement learning to AI was “Theory of Neural-Analog Reinforcement Systems”[3] by Minsky, 1954 A cool reinforcement learning demo is the DeepMind Atari video from last year[4] [1] [LINK] [2] [LIN… — paulsutter, 2015-10
Thanks for the correction. The "arbitrary" qualifier is not in TFA, but (as, indeed, you said) that's the point of the demo, e.g.: [VIDEO] Note that they're using just the video signal from the game as input. It's really a sad comment on the state of reporting at MIT Tech Review that you learn more about the tech from a youtube video than from an article. (My complaint is not with the DeepMind pe… — mturmon, 2014-12
AGI is the idea of creating general intelligence. Which almost certainly requires some reinforcement learning (the ways dogs and humans learn, by having a goal and an internal reward system). Deepmind's Atari player is based on deep reinforcement learning, where increasing the score represents a reward: [VIDEO] I used to believe the moving goalpost idea, that AI is anything that isn't yet possibl… — paulsutter, 2016-01
I was also under the misapprehension that deep learning is just about classification, but that isn't true. Yes, reinforcement learning is the path to general intelligence, and the deep learning community is showing impressive progress on that front as well. The Deepmind demo [1] and the recent robotics work at Berkeley[2] are good examples. Thanks for the link to Stan Franklin's work. I'm glad to… — paulsutter, 2015-05
Related videos
- Tom Lehrer - New Math (Animated) · Jared Khan · 4:23
- GPUs to Mars: Full-Scale Simulation of SpaceX's Mars Rocket Engine · insideHPC Report · 46:36
- Erlang The Movie II: The Sequel · gar1t · 9:55
- Tensorflow and deep learning - without a PhD by Martin Görner · Devoxx · 2:35:53
- The spelled-out intro to neural networks and backpropagation: building micrograd · Andrej Karpathy · 2:25:52
- But what is a neural network? | Deep learning chapter 1 · 3Blue1Brown · 18:40
- Overwatch Gameplay Architecture and Netcode · GDC Festival of Gaming · 1:04:03
- AlphaGo - The Movie | Full award-winning documentary · Google DeepMind · 1:30:28
- Transformers, the tech behind LLMs | Deep Learning Chapter 5 · 3Blue1Brown · 27:14
- USENIX Security '18-Q: Why Do Keynote Speakers Keep Suggesting That Improving Security Is Possible? · USENIX · 51:22
- Lecture 1 | Machine Learning (Stanford) · Stanford · 1:08:40
- RustConf 2018 - Closing Keynote - Using Rust For Game Development by Catherine West · Rust Programming Language · 42:29
- A Deep Dive into Nanite Virtualized Geometry · SIGGRAPH Advances in Real-Time Rendering · 1:09:59
- How Crash Bandicoot Hacked The Original Playstation | War Stories | Ars Technica · Ars Technica · 31:37
- MarI/O - Machine Learning for Video Games · SethBling · 5:57
- Deep Dive into LLMs like ChatGPT · Andrej Karpathy · 3:31:23