Add to ORKGDeep Reinforcement Learning (RL) agents often overfit the training environment, leading to poor generalization performance. In this paper, we propose Thinker, a bootstrapping method to remove adversarial effects of confounding features from the observation in an unsupervised way, and thus, it improves RL agents' generalization. Thinker first clusters experience trajectories into several clusters. These trajectories are then bootstrapped by applying a style transfer generator, which translates the trajectories from one cluster's style to another while maintaining the content of the observations. The bootstrapped trajectories are then used for policy learning. Thinker has wide applicability among many RL settings. Experimental results reveal ...
Skill-based reinforcement learning (RL) has emerged as a promising strategy to leverage prior knowle...
In recent years, the advances in robotics have allowed for robots to venture into places too dangero...
We present an algorithm for Inverse Reinforcement Learning (IRL) from expert state observations only...
A machine learning (ML) system must learn not only to match the output of a target function on a tra...
Many important tasks are defined in terms of object. To generalize across these tasks, a reinforceme...
In this thesis we aim to improve generalisation in deep reinforcement learning. Generalisation is a ...
Reinforcement Learning (RL) has seen exponential performance improvements over the past decade, achi...
Adapting a Reinforcement Learning (RL) agent to an unseen environment is a difficult task due to typ...
The development of reinforcement learning attracts more and more attention among researchers. Levera...
Typically in reinforcement learning, agents are trained and evaluated on the same environment. Conse...
Different from classic Supervised Learning, Reinforcement Learning (RL), is fundamentally interactiv...
We propose a method for meta-learning reinforcement learning algorithms by searching over the space ...
Learning a policy with great generalization to unseen environments remains challenging but critical ...
In control, the objective is to find a mapping from states to actions that steer a system to a desir...
Recent progress in reinforcement learning (RL) has started producing generally capable agents that c...
Skill-based reinforcement learning (RL) has emerged as a promising strategy to leverage prior knowle...
In recent years, the advances in robotics have allowed for robots to venture into places too dangero...
We present an algorithm for Inverse Reinforcement Learning (IRL) from expert state observations only...
A machine learning (ML) system must learn not only to match the output of a target function on a tra...
Many important tasks are defined in terms of object. To generalize across these tasks, a reinforceme...
In this thesis we aim to improve generalisation in deep reinforcement learning. Generalisation is a ...
Reinforcement Learning (RL) has seen exponential performance improvements over the past decade, achi...
Adapting a Reinforcement Learning (RL) agent to an unseen environment is a difficult task due to typ...
The development of reinforcement learning attracts more and more attention among researchers. Levera...
Typically in reinforcement learning, agents are trained and evaluated on the same environment. Conse...
Different from classic Supervised Learning, Reinforcement Learning (RL), is fundamentally interactiv...
We propose a method for meta-learning reinforcement learning algorithms by searching over the space ...
Learning a policy with great generalization to unseen environments remains challenging but critical ...
In control, the objective is to find a mapping from states to actions that steer a system to a desir...
Recent progress in reinforcement learning (RL) has started producing generally capable agents that c...
Skill-based reinforcement learning (RL) has emerged as a promising strategy to leverage prior knowle...
In recent years, the advances in robotics have allowed for robots to venture into places too dangero...
We present an algorithm for Inverse Reinforcement Learning (IRL) from expert state observations only...