18 citations · 28 across the 14 of their papers we have counts for
15 papers
Evaluating RL Explainability Methods by How Much They Help Fix Bugs in Agents
Ram Rachum, Yotam Amitai, Bálint Gyevnár +2
This preliminary paper outlines a planned evaluation benchmark for Explainable Reinforcement Learning (XRL) methods. Current evaluations rely on functionally-grounded metrics like…
BXRL: Behavior-Explainable Reinforcement Learning
Ram Rachum, Yotam Amitai, Yonatan Nakar +2
A major challenge of Reinforcement Learning is that agents often learn undesired behaviors that seem to defy the reward structure they were given. Explainable Reinforcement Learnin…
Truthfulness Despite Weak Supervision: Evaluating and Training LLMs Using Peer Prediction
Tianyi Alex Qiu, Micah Carroll, Cameron Allen
The evaluation and post-training of large language models (LLMs) rely on supervision, but strong supervision for difficult tasks is often unavailable, especially when evaluating fr…
Focused Skill Discovery: Learning to Control Specific State Variables while Minimizing Side Effects
Jonathan Colaço Carr, Qinyi Sun, Cameron Allen
Skills are essential for unlocking higher levels of problem solving. A common approach to discovering these skills is to learn ones that reliably reach different states, thus empow…
From Pixels to Factors: Learning Independently Controllable State Variables for Reinforcement Learning
Rafael Rodriguez-Sanchez, Cameron Allen, George Konidaris
Algorithms that exploit factored Markov decision processes are far more sample-efficient than factor-agnostic methods, yet they assume a factored representation is known a priori -…
Benchmarking Partial Observability in Reinforcement Learning with a Suite of Memory-Improvable Domains
Ruo Yu Tao, Kaicheng Guo, Cameron Allen +1
Mitigating partial observability is a necessary but challenging task for general reinforcement learning algorithms. To improve an algorithm's ability to mitigate partial observabil…
Mitigating Partial Observability in Sequential Decision Processes via the Lambda Discrepancy
Cameron Allen, Aaron Kirtland, Ruo Yu Tao +7
Reinforcement learning algorithms typically rely on the assumption that the environment dynamics and value function can be expressed in terms of a Markovian state representation. H…
Evidence of Learned Look-Ahead in a Chess-Playing Neural Network
Erik Jenner, Shreyas Kapur, Vasil Georgiev +3
Do neural networks learn to implement algorithms such as look-ahead or search "in the wild"? Or do they rely purely on collections of simple heuristics? We present evidence of lear…
Characterizing the Action-Generalization Gap in Deep Q-Learning
Zhiyuan Zhou, Cameron Allen, Kavosh Asadi +1
We study the action generalization ability of deep Q-learning in discrete action spaces. Generalization is crucial for efficient reinforcement learning (RL) because it allows agent…
Coarse-Grained Smoothness for RL in Metric Spaces
Omer Gottesman, Kavosh Asadi, Cameron Allen +3
Principled decision-making in continuous state--action spaces is impossible without some assumptions. A common approach is to assume Lipschitz continuity of the Q-function. We show…
Bad-Policy Density: A Measure of Reinforcement Learning Hardness
David Abel, Cameron Allen, Dilip Arumugam +3
Reinforcement learning is hard in general. Yet, in many specific environments, learning is easy. What makes learning easy in one environment, but difficult in another? We address t…
Learning Markov State Abstractions for Deep Reinforcement Learning
Cameron Allen, Neev Parikh, Omer Gottesman +1
A fundamental assumption of reinforcement learning in Markov decision processes (MDPs) is that the relevant decision process is, in fact, Markov. However, when MDPs have rich obser…
Task Scoping: Generating Task-Specific Abstractions for Planning in Open-Scope Models
Michael Fishman, Nishanth Kumar, Cameron Allen +4
A general-purpose planning agent requires an open-scope world model: one rich enough to tackle any of the wide range of tasks it may be asked to solve over its operational lifetime…
Efficient Black-Box Planning Using Macro-Actions with Focused Effects
Cameron Allen, Michael Katz, Tim Klinger +3
The difficulty of deterministic planning increases exponentially with search-tree depth. Black-box planning presents an even greater challenge, since planners must operate without…
Mean Actor Critic
Cameron Allen, Kavosh Asadi, Melrose Roderick +3
We propose a new algorithm, Mean Actor-Critic (MAC), for discrete-action continuous-state reinforcement learning. MAC is a policy gradient algorithm that uses the agent's explicit…