Research map

Reinforcement learning: research map

Policy learning, offline RL, imitation and learning from feedback.

6,108 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026), in 6 topics. Within all five venues, its share held steady from 5.9% in 2023–24 to 5.1% in 2025–26 (1,568 → 2,317 papers at ICML, NeurIPS, CVPR and AAAI, the venues with data for all four years).

Explore Reinforcement learning in the interactive map

Topics

• Visual and representation RL · 2,133 papers
  In-context and adaptive RL, Scalable RL systems and environments, Probabilistic and autonomous RL
• Policy gradient and actor-critic · 1,502 papers
  Model-based RL with guarantees, Regularized scalable deep value learning, Temporal-difference and discounting theory
• RL from feedback · 737 papers
  RL for LLM reasoning, Scaling RL with verifiable rewards, GRPO stability and variants
• MDPs and planning · 682 papers
  Average-reward and regularized MDPs, POMDPs and confounded RL, Robust MDP policy gradient
• Offline RL · 623 papers
  Offline RL sample complexity theory, Simple practical offline RL, Pessimism and uncertainty in offline RL
• Imitation learning · 431 papers
  Offline imitation from imperfect data, Imitation combined with RL, Behavior cloning theory and models

Most cited papers in Reinforcement learning

• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor (ICML 2018 · 3,754 citations)
• Addressing Function Approximation Error in Actor-Critic Methods (ICML 2018 · 2,316 citations)
• Self-Critical Sequence Training for Image Captioning (CVPR 2017 · 2,268 citations)
• Dueling Network Architectures for Deep Reinforcement Learning (ICML 2016 · 1,777 citations)
• A Deep Reinforced Model for Abstractive Summarization (ICLR 2018 · 1,255 citations)
• Benchmarking Deep Reinforcement Learning for Continuous Control (ICML 2016 · 949 citations)
• Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation (NeurIPS 2016 · 822 citations)
• Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization (ICML 2016 · 591 citations)
• Action-Decision Networks for Visual Tracking With Deep Reinforcement Learning (CVPR 2017 · 574 citations)
• Conservative Q-Learning for Offline Reinforcement Learning (NeurIPS 2020 · 556 citations)
• Deep Reinforcement Learning from Human Preferences (NeurIPS 2017 · 500 citations)
• Decision Transformer: Reinforcement Learning via Sequence Modeling (NeurIPS 2021 · 464 citations)