6 papers
General Flexible -divergence for Challenging Offline RL Datasets with Low Stochasticity and Diverse Behavior Policies
Jianxun Wang, Grant C. Forbes, Leonardo Villalobos-Arias +1
Offline RL algorithms aim to improve upon the behavior policy that produces the collected data while constraining the learned policy to be within the support of the dataset. Howeve…
Minding Motivation: The Effect of Intrinsic Motivation on Agent Behaviors
Leonardo Villalobos-Arias, Grant Forbes, Jianxun Wang +2
Games are challenging for Reinforcement Learning~(RL) agents due to their reward-sparsity, as rewards are only obtainable after long sequences of deliberate actions. Intrinsic Moti…
Action-Dependent Optimality-Preserving Reward Shaping
Grant C. Forbes, Jianxun Wang, Leonardo Villalobos-Arias +2
Recent RL research has utilized reward shaping--particularly complex shaping rewards such as intrinsic motivation (IM)--to encourage agent exploration in sparse-reward environments…
Potential-Based Intrinsic Motivation: Preserving Optimality With Complex, Non-Markovian Shaping Rewards
Grant C. Forbes, Leonardo Villalobos-Arias, Jianxun Wang +2
Recently there has been a proliferation of intrinsic motivation (IM) reward-shaping methods to learn in complex and sparse-reward environments. These methods can often inadvertentl…
Potential-Based Reward Shaping For Intrinsic Motivation
Grant C. Forbes, Nitish Gupta, Leonardo Villalobos-Arias +3
Recently there has been a proliferation of intrinsic motivation (IM) reward-shaping methods to learn in complex and sparse-reward environments. These methods can often inadvertentl…
Modeling Risk in Reinforcement Learning: A Literature Mapping
Leonardo Villalobos-Arias, Derek Martin, Abhijeet Krishnan +3
Safe reinforcement learning deals with mitigating or avoiding unsafe situations by reinforcement learning (RL) agents. Safe RL approaches are based on specific risk representations…