SOTAVerified

Reinforcement Learning (RL)

Reinforcement Learning (RL) involves training an agent to take actions in an environment to maximize a cumulative reward signal. The agent interacts with the environment and learns by receiving feedback in the form of rewards or punishments for its actions. The goal of reinforcement learning is to find the optimal policy or decision-making strategy that maximizes the long-term reward.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 4141–4150 of 15113 papers

TitleStatusHype
Reinforcement Learning with Generalizable Gaussian Splatting—0
Pessimistic Causal Reinforcement Learning with Mediators for Confounded Offline Data—0
Prior-dependent analysis of posterior sampling reinforcement learning with function approximation—0
Independent RL for Cooperative-Competitive Agents: A Mean-Field Perspective—0
Causality from Bottom to Top: A Survey—0
Distributed Multi-Objective Dynamic Offloading Scheduling for Air-Ground Cooperative MEC—0
Neural-Kernel Conditional Mean Embeddings—0
The Fallacy of Minimizing Cumulative Regret in the Sequential Task Setting—0
ViSaRL: Visual Reinforcement Learning Guided by Human Saliency—0
EXPLORER: Exploration-guided Reasoning for Textual Reinforcement LearningCode0
Show:102550
← PrevPage 415 of 1512Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1PPGMean Normalized Performance0.76—Unverified
2PPOMean Normalized Performance0.58—Unverified