Latest Articles for Reinforcement Learning

SAC-BBF combines SAC and Rainbow for better learning in discrete action spaces.

Sequential policies improve performance in dynamic algorithm configuration with interrelated choices.

Examining the impact of periodic activation functions on learning efficiency and generalization.

A new framework enhances learning efficiency in reinforcement tasks.

A new method enhances robot exploration and goal achievement through temporal distance.

An overview of experience replay methods and their impact on learning efficiency.

This article introduces ValueWalk, a method for improving computer learning from expert behavior.

A new framework boosts agent learning by focusing on causal relationships.

Introducing Random Latent Exploration: a novel approach to improve agent exploration.

Introducing PG-Rainbow, a new algorithm enhancing decision-making in AI.

This study evaluates the impact of visual feature extractors on robot learning tasks.

Gymnasium simplifies research in reinforcement learning with versatile environments and support tools.

A technique to enhance training datasets for better agent performance.

New methods enhance speed and stability in value iteration.

ARCLE helps researchers improve AI's problem-solving skills using reinforcement learning.

New policies improve decision-making in artificial intelligence through flexible exploration.

A new approach enhances agent learning with flexible reward systems.

Enhancing Policy Mirror Descent with functional acceleration for faster decision-making.

A look at improving train scheduling with continual learning agents.

Discover the new Morse Model-based RL approach for effective decision-making.

A new approach improves decision-making in reinforcement learning by focusing on objects.

SUMO enhances decision-making in offline reinforcement learning by improving uncertainty estimation.

A fresh approach to improving learning amid noise in reinforcement learning environments.

Research unveils advanced settings for observing language growth among artificial agents.

An in-depth look at the implications of centralized critics in MARL.

A fresh approach improves error modeling, enhancing decision-making in complex environments.

Examining no regret strategies in two-player Stackelberg games for improved decision-making.

Combining optimization strategies improves learning in reinforcement learning tasks.

DuelMIX improves cooperation among agents in multi-agent reinforcement learning environments.

This article explores effective training methods for reinforcement learning agents to reduce regret.

An overview of the UCB algorithm and its stability in data collection.

A new framework enhances learning efficiency in visual reinforcement learning.

A new approach improves actor-critic methods in reinforcement learning.

New methods improve safe exploration in reinforcement learning systems.

An overview of CTDE methods in multi-agent reinforcement learning.

A new model improves AI's ability to learn without forgetting.

Exploring a new method for Q-function approximation in reinforcement learning.

A new method enhances learning performance through dynamic action persistence.

This study examines the impact of regularization methods on actor networks in offline RL.

Exploring vulnerabilities of cooperative multi-agent systems to backdoor attacks.