Best Reinforcement Learning Papers

The highest-signal papers on Reinforcement Learning, ranked by community reviews and momentum.
Canonical intent: topic=reinforcement-learning|type=paper|year=evergreen

Explore TopicAwesome ListsResearch Atlas

Top Picks

8
KARL: Knowledge Agents via Reinforcement Learning
Jonathan D. Chang, Andrew Drozdov, Shubham Toshniwal, Owen Oertell, Alexander Trott, Jacob Portes, Abhay Gupta, Pallavi Koppol, Ashutosh Baheti, Sean Kulinski, Ivan Zhou, Irene Dea, Krista Opsahl-Ong, Simon Favreau-Lessard, Sean Owen, Jose Javier Gonzalez Ortiz, Arnav Singhvi, Xabi Andrade, Cindy Wang, Kartik Sreenivasan, Sam Havens, Jialu Liu, Peyton DeNiro, Wen Sun, Michael Bendersky, Jonathan Frankle
Mar 5, 2026·363 checkouts·arxiv.org
Source ↗
13
A Generalist Agent
Scott Reed, Konrad Zolna, Emilio Parisotto, Sergio Gomez Colmenarejo, Nando de Freitas
May 12, 2022·325 checkouts·arxiv.org
Source ↗
21
Mastering the game of Go without human knowledge
David Silver, Julian Schrittwieser, Karen Simonyan, Ioannis Antonoglou, Aja Huang, Arthur Guez, Thomas Hubert, Lucas Baker, Matthew Lai, Adrian Bolton, Yutian Chen, Timothy Lillicrap, Fan Hui, Laurent Sifre, George van den Driessche, Thore Graepel, Demis Hassabis
Oct 18, 2017·227 checkouts·doi.org
Source ↗
27
Hindsight Experience Replay
Marcin Andrychowicz, Filip Wolski, Alex Ray, Jonas Schneider, Rachel Fong, Peter Welinder, Bob McGrew, Josh Tobin, Pieter Abbeel, Wojciech Zaremba
Jul 5, 2017·140 checkouts·arxiv.org
Source ↗

FAQ

How is this “best Reinforcement Learning Papers” collection ranked?

This page ranks Reinforcement Learning Papers using topic relevance, checkout momentum, source diversity, and freshness signals. Rankings are recalculated as new items and engagement arrive, so readers see resources that are both high quality and currently useful for implementation, research, and practical decision making. Canonical intent key: topic=reinforcement-learning|type=paper|year=evergreen.

How do you prevent duplicate collection pages?

Attendemia maps each slug variant, including best-of and year forms, to one canonical intent key. If two URLs describe the same topic, type, and timeframe, non-canonical versions permanently redirect. This consolidates crawl signals, avoids duplicate content dilution, and helps search engines index the strongest single page.

When does a year page stay separate from evergreen?

A year-specific page stays separate only when its item set is materially different from evergreen and has enough ranking depth. When overlap is high, the year URL redirects to the evergreen canonical page. This avoids thin duplication while preserving genuinely distinct annual collections for search users.

Are these paid recommendations?

No. These recommendations are not paid placements. Attendemia ranks items from public metadata, source quality coverage, and user engagement signals, then orders them by practical usefulness. Sponsorship does not buy rank position, so this page should be interpreted as editorial curation rather than advertising inventory.