{"id":"arxiv:1509.06461","title":"Deep Reinforcement Learning with Double Q-learning","date":"2015-09-22","section":"Reinforcement learning","quality":0.079916,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/","source_url":"https://arxiv.org/abs/1509.06461","text_available":true,"authors":["Hado van Hasselt","Arthur Guez","David Silver"],"outgoing":[],"incoming":[{"source":"arxiv:1511.05952","target":"arxiv:1509.06461","evidence":"id","id":"arxiv:1511.05952","title":"Prioritized Experience Replay","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/c88bf0c11c1817dcc457c001dfef5cd120a0946c8a009a0fe40648dae03eb7d4/"},{"source":"arxiv:1511.06581","target":"arxiv:1509.06461","evidence":"id","id":"arxiv:1511.06581","title":"Dueling Network Architectures for Deep Reinforcement Learning","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/0a5f0593cf943c782ba38ae2fc386f953f859b9d3fc9722e8f17f67ca9cf50b5/"},{"source":"arxiv:1707.06887","target":"arxiv:1509.06461","evidence":"title","id":"arxiv:1707.06887","title":"A Distributional Perspective on Reinforcement Learning","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/96b040bb771d28ca3254b5519a2191c3e958a35dfdd60075344ac86af6a590e1/"},{"source":"arxiv:1710.02298","target":"arxiv:1509.06461","evidence":"title","id":"arxiv:1710.02298","title":"Rainbow: Combining Improvements in Deep Reinforcement Learning","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/4183a5aecdb6558733bd0b2028e6197904c8b46173181bf6c8270fb6cbcfc973/"},{"source":"arxiv:1712.06567","target":"arxiv:1509.06461","evidence":"title","id":"arxiv:1712.06567","title":"Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/7191dc11e224beb3dc9abb8cd4efc58c70b79939b61fd56e4416b4b5e1404312/"},{"source":"arxiv:1802.09477","target":"arxiv:1509.06461","evidence":"title","id":"arxiv:1802.09477","title":"Addressing Function Approximation Error in Actor-Critic Methods","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/5d1a410c86d85da493735fae67eb1b2689350e5ac449d39717deafaf3904e7c1/"},{"source":"arxiv:1910.01708","target":"arxiv:1509.06461","evidence":"title","id":"arxiv:1910.01708","title":"Benchmarking Batch Deep Reinforcement Learning Algorithms","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/ce9cb9b764eb7fdaf1aa766cb24e6abaec20a00445ad5af5a6dacb5af3ad3ddc/"},{"source":"arxiv:2011.14826","target":"arxiv:1509.06461","evidence":"id","id":"arxiv:2011.14826","title":"Revisiting Rainbow: Promoting more Insightful and Inclusive Deep Reinforcement Learning Research","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/2bf20803a5d9e686543fe5f3967e69b5aa481ba29ff91c9ca3fbeb147ab8c0aa/"},{"source":"arxiv:2205.07802","target":"arxiv:1509.06461","evidence":"title","id":"arxiv:2205.07802","title":"The Primacy Bias in Deep Reinforcement Learning","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/4e2aca370b941bb16bb6fefc359c420f6d4bde5a956bedaf0a84e6e71febc5bc/"},{"source":"arxiv:2305.19452","target":"arxiv:1509.06461","evidence":"id","id":"arxiv:2305.19452","title":"Bigger, Better, Faster: Human-level Atari with human-level efficiency","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/727a7ee43fd005401d34d4591a2f49d85175f9366ebf2c9e7a862b340787614e/"},{"source":"arxiv:2411.03820","target":"arxiv:1509.06461","evidence":"title","id":"arxiv:2411.03820","title":"Beyond The Rainbow: High Performance Deep Reinforcement Learning on a Desktop PC","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/8bcc96ef608c017d3ddb7ded7dc9e3b760390f7df441214eeee08a2586c981ea/"},{"source":"doi:10.1073/pnas.1611835114","target":"arxiv:1509.06461","evidence":"title","id":"doi:10.1073/pnas.1611835114","title":"Overcoming catastrophic forgetting in neural networks","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/849082368a44d06b93ef3ba95bcfddc6672b7a5e8064ad2b2d19601639266e8b/"}],"text_parts":[{"start":0,"end":15427,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-1.html"},{"start":15427,"end":30884,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html"},{"start":30884,"end":40120,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-3.html"}],"contents":[{"title":"Deep Reinforcement Learning with Double Q-learning","offset":0,"level":1,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-1.html#text"},{"title":"Abstract","offset":54,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-1.html#text"},{"title":"Background","offset":4188,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-1.html#text"},{"title":"Deep Q Networks","offset":6178,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-1.html#text"},{"title":"Double Q-learning","offset":7462,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-1.html#text"},{"title":"Overoptimism due to estimation errors","offset":9328,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-1.html#text"},{"title":"Theorem 1 .","offset":10736,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-1.html#text"},{"title":"Double DQN","offset":18251,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Empirical results","offset":19766,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Results on overoptimism","offset":21214,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Quality of the learned policies","offset":23731,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Robustness to Human starts","offset":26226,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Discussion","offset":28919,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Acknowledgments","offset":29789,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Appendix","offset":30023,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Theorem 1 .","offset":30036,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Proof of Theorem 1.","offset":30856,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-2.html#text"},{"title":"Theorem 2 .","offset":34187,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-3.html#text"},{"title":"Proof.","offset":34650,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-3.html#text"},{"title":"Experimental Details for the Atari 2600 Domain","offset":37104,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-3.html#text"},{"title":"Network Architecture","offset":37490,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-3.html#text"},{"title":"Hyper-parameters","offset":38321,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-3.html#text"},{"title":"Supplementary Results in the Atari 2600 Domain","offset":39013,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-3.html#text"},{"title":"Instructions for reporting errors","offset":39156,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/part-3.html#text"}],"text_code_points":40120,"source_markdown":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/a2f115ac879cd2efb9548baf3ecdcb925d2492f03b4b0940720f6e5b0840715b/source.md"}