A Taxonomy of Exploration RL Methods

• Tianpei Yang et al
• 2
• 3
• 5
• 6
• 7
• 8
• 9
• 11
• 12
• 13
• 16
• 17
• 18
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 4
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 1
• 4
• 15
• 16

Papers

Classic Exploration RL Papers

• First return then explore
• \#Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
• VIME: Variational information maximizing exploration
• Deep Exploration via Bootstrapped DQN
• (More) Efficient Reinforcement Learning via Posterior Sampling
• Using Confidence Bounds for Exploitation-Exploration Trade-offs
• How can we define intrinsic motivation?
• A Contextual-Bandit Approach to Personalized News Article Recommendation
• An empirical evaluation of thompson sampling
• A Tutorial on Thompson Sampling
• Large-Scale Study of Curiosity-Driven Learning
• Episodic Curiosity through Reachability
• Self-Supervised Exploration via Disagreement
• EMI: Exploration with Mutual Information
• Optimistic Exploration even with a Pessimistic Initialisation
• RIDE: Rewarding Impact-Driven Exploration for Procedurally-Generated Environments
• Neural Contextual Bandits with UCB-based Exploration
• MNIST
• Rank the Episodes: A Simple Approach for Exploration in Procedurally-Generated Environments
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• First return then explore
• Atari
• rllab
• A Tutorial on Thompson Sampling
• VizDoom
• Optimistic Exploration even with a Pessimistic Initialisation
• MNIST
• Neural Contextual Bandits with UCB-based Exploration

ICLR 2022

• multi-armed bandit
• The Information Geometry of Unsupervised Reinforcement Learning
• When should agents explore?
• Learning more skills through optimistic exploration
• Learning Long-Term Reward Redistribution via Randomized Return Decomposition
• Reinforcement Learning with Sparse Rewards using Guidance from Offline Demonstration
• Generative Planning for Temporally Coordinated Exploration in Reinforcement Learning
• Learning Altruistic Behaviours in Reinforcement Learning without External Rewards
• Anti-Concentrated Confidence Bonuses for Scalable Exploration
• Lipschitz-constrained Unsupervised Skill Discovery
• LIGS: Learnable Intrinsic-Reward Generation Selection for Multi-Agent Learning
• Multi-Stage Episodic Control for Strategic Exploration in Text Games
• On the Convergence of the Monte Carlo Exploring Starts Algorithm for Reinforcement Learning
• blackjack
• foraging
• Jericho
• LIGS: Learnable Intrinsic-Reward Generation Selection for Multi-Agent Learning

ICLR 2023

• Learnable Behavior Control: Breaking Atari Human World Records via Sample-Efficient Behavior Selection
• The Role of Coverage in Online Reinforcement Learning
• Near-optimal Policy Identification in Active Reinforcement Learning
• Planning Goals for Exploration
• Pink Noise Is All You Need: Colored Noise Exploration in Deep Reinforcement Learning
• Learning About Progress From Experts
• DEP-RL: Embodied Exploration for Reinforcement Learning in Overactuated and Musculoskeletal Systems
• Does Zero-Shot Reinforcement Learning Exist?
• Human-level Atari 200x faster
• Learning Achievement Structure for Structured Exploration in Domains with Sparse Reward
• Safe Exploration Incurs Nearly No Additional Sample Complexity for Reward-Free RL
• Latent State Marginalization as a Low-cost Approach to Improving Exploration
• Revisiting Curiosity for Exploration in Procedurally Generated Environments
• MoDem: Accelerating Visual Model-Based Reinforcement Learning with Demonstrations
• Simplifying Model-based RL: Learning Representations, Latent-space Models, and Policies with One Objective
• EUCLID: Towards Efficient Unsupervised Reinforcement Learning with Multi-choice Dynamics Model
• Guarded Policy Optimization with Imperfect Online Demonstrations
• NetHack
• DeepMind Control Suite

ICLR 2024

• A Theoretical Explanation of Deep RL Performance in Stochastic Environments
• DrM: Mastering Visual Reinforcement Learning through Dormant Ratio Minimization
• METRA: Scalable Unsupervised RL with Metric-Aware Abstraction
• Text2Reward: Reward Shaping with Language Models for Reinforcement Learning
• Pre-Training Goal-based Models for Sample-Efficient Reinforcement Learning