Hacker News
- Deep reinforcement learning doesn't work yet https://www.alexirpan.com/2018/02/14/rl-hard.html 50 comments
- Deep Reinforcement Learning Doesn't Work (yet) https://www.alexirpan.com/2018/02/14/rl-hard.html 2 comments
- Deep Reinforcement Learning Doesn't Work Yet. Posted in 2018. Six years later, how much have things changed and what remained the same in your opinion? https://www.alexirpan.com/2018/02/14/rl-hard.html 24 comments reinforcementlearning
- Deep Reinforcement Learning Doesn't Work Yet (Feb 2018) https://www.alexirpan.com/2018/02/14/rl-hard.html 3 comments reinforcementlearning
- "Deep Reinforcement Learning Doesn't Work Yet": sample-inefficient, outperformed by domain-specific models or techniques, fragile reward functions, gets stuck in local optima, unreproducible & undebuggable, & doesn't generalize https://www.alexirpan.com/2018/02/14/rl-hard.html 9 comments reinforcementlearning
Linking pages
- Lessons Learned Reproducing a Deep Reinforcement Learning Paper http://amid.fish/reproducing-deep-rl 37 comments
- Strong AI Requires Autonomous Building of Composable Models https://thegradient.pub/strong-ai-requires-autonomous-building-of-composable-models/ 33 comments
- Reinforcement Learning from scratch | by Emmanuel Ameisen | Insight https://blog.insightdatascience.com/reinforcement-learning-from-scratch-819b65f074d8 25 comments
- Deep Reinforcement Learning Works - Now What? • Chen Tessler https://tesslerc.github.io/posts/drl_works_now_what/ 19 comments
- RAdam: A New State-of-the-Art Optimizer for RL? | by Chris Nota | Autonomous Learning Library | Medium https://medium.com/autonomous-learning-library/radam-a-new-state-of-the-art-optimizer-for-rl-442c1e830564 10 comments
- Marvel, Stanford & CMU NLP Playlists, Voynich, Bitter Lesson Vol. 2, ICLR 2019, Dialogue Demos | Revue http://newsletter.ruder.io/issues/marvel-stanford-cmu-nlp-playlists-voynich-bitter-lesson-vol-2-iclr-2019-dialogue-demos-173432 1 comment
- Applications of Reinforcement Learning in Real World | by Gary Chan | Towards Data Science https://towardsdatascience.com/applications-of-reinforcement-learning-in-real-world-1a94955bcd12 0 comments
- Does Hierarchial Reinforcement Learning work yet? – Sholto's Blog – A collection of experiments https://sholtodouglas.github.io/DoesHierarchialRLWorkYet/ 0 comments
- Reinforcement Learning: Playing Doom with PyTorch https://brandonlmorris.com/2018/10/09/dql-vizdoom/ 0 comments
- Reinforcement Learning: A Deep Dive | Toptal https://www.toptal.com/machine-learning/deep-dive-into-reinforcement-learning 0 comments
- GitHub - adeshpande3/Machine-Learning-Links-And-Lessons-Learned: List of all the lessons learned, best practices, and links from my time studying machine learning https://github.com/adeshpande3/Machine-Learning-Links-And-Lessons-Learned 0 comments
- GitHub - SoyGema/Startcraft_pysc2_minigames: Startcraft II Machine Learning research with DeepMind pysc2 python library .mini-games and agents. https://github.com/SoyGema/Startcraft_pysc2_minigames 0 comments
- Avoiding Moravec’s Paradox Solves AGI in Five Years | by Carlos E. Perez | Intuition Machine | Medium https://medium.com/intuitionmachine/near-term-agi-should-be-considered-as-a-possibility-9bcf276f9b16 0 comments
- The evolution of intelligence in robots: Part 2 | by Simon Kalouche | DataDrivenInvestor https://medium.com/datadriveninvestor/the-evolution-of-intelligence-in-robots-part-2-43aea6985b8f 0 comments
- Machine Learning — The New Programming Language | by Raul Incze | Cognifeed | Medium https://medium.com/cognifeed/a-view-on-machine-learning-4b6118940db4 0 comments
- Generating Natural-Language Text with Neural Networks | by Jonathan Mugan | Medium https://medium.com/@jmugan/generating-natural-language-text-with-neural-networks-e983bb48caad 0 comments
- A primer on my favorite pessimistic scientific articles https://www.abhishaike.com/p/a-primer-on-my-favorite-pessimistic 0 comments
- Implementing Deep Q-Network: a Reinforcement Learning Beginner's Challenges and Learnings https://mingfei.io/dqn/ 0 comments
Linked pages
- Zayd's Blog – Why is machine learning 'hard'? http://ai.stanford.edu/~zayd/why-is-machine-learning-hard.html 307 comments
- How to assign partial credit on an exam of true-false questions? | What's new https://terrytao.wordpress.com/2016/06/01/how-to-assign-partial-credit-on-an-exam-of-true-false-questions/ 242 comments
- Competitive Self-Play https://blog.openai.com/competitive-self-play/ 138 comments
- How HBO’s Silicon Valley built “Not Hotdog” with mobile TensorFlow, Keras & React Native | by Tim Anglade | Medium https://medium.com/@timanglade/how-hbos-silicon-valley-built-not-hotdog-with-mobile-tensorflow-keras-react-native-ef03260747f3 136 comments
- [1701.01724] DeepStack: Expert-Level Artificial Intelligence in No-Limit Poker https://arxiv.org/abs/1701.01724 87 comments
- Dota 2 https://blog.openai.com/dota-2/ 87 comments
- [1702.06230] Beating the World's Best at Super Smash Bros. with Deep Reinforcement Learning https://arxiv.org/abs/1702.06230 56 comments
- [1611.01578] Neural Architecture Search with Reinforcement Learning https://arxiv.org/abs/1611.01578 38 comments
- Enabling Continual Learning in Neural Networks https://deepmind.com/blog/enabling-continual-learning-in-neural-networks/ 33 comments
- GitHub - mgbellemare/Arcade-Learning-Environment: The Arcade Learning Environment (ALE) -- a platform for AI research. https://github.com/mgbellemare/Arcade-Learning-Environment 21 comments
- Closing the Simulation-to-Reality Gap for Deep Robotic Learning – Google AI Blog http://research.googleblog.com/2017/10/closing-simulation-to-reality-gap-for.html 12 comments
- [1707.01495] Hindsight Experience Replay https://arxiv.org/abs/1707.01495 11 comments
- https://arxiv.org/pdf/1707.06887.pdf 10 comments
- Learning from Human Preferences https://blog.openai.com/deep-reinforcement-learning-from-human-preferences/ 7 comments
- [1611.05397] Reinforcement Learning with Unsupervised Auxiliary Tasks https://arxiv.org/abs/1611.05397 6 comments
- Model-based Reinforcement Learning with Neural Network Dynamics – The Berkeley Artificial Intelligence Research Blog http://bair.berkeley.edu/blog/2017/11/30/model-based-rl/ 6 comments
- Spam detection in the physical world https://blog.openai.com/spam-detection-in-the-physical-world/ 6 comments
- [1711.00832] A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning https://arxiv.org/abs/1711.00832 5 comments
- Nuts and Bolts of Applying Deep Learning (Andrew Ng) - YouTube https://www.youtube.com/watch?v=F1ka6a13S9I 4 comments
- Human-level control through deep reinforcement learning | Nature https://www.nature.com/articles/nature14236 3 comments
Related searches:
Search whole site: site:alexirpan.com
Search title: Deep Reinforcement Learning Doesn't Work Yet
See how to search.