Inverse Reinforcement Learning from Non-Stationary Learning Agents

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Sivakumar, Kavinayan P., Shen, Yi, Bell, Zachary, Nivison, Scott, Chen, Boyuan, Zavlanos, Michael M.
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913553032675328
author Sivakumar, Kavinayan P.
Shen, Yi
Bell, Zachary
Nivison, Scott
Chen, Boyuan
Zavlanos, Michael M.
author_facet Sivakumar, Kavinayan P.
Shen, Yi
Bell, Zachary
Nivison, Scott
Chen, Boyuan
Zavlanos, Michael M.
contents In this paper, we study an inverse reinforcement learning problem that involves learning the reward function of a learning agent using trajectory data collected while this agent is learning its optimal policy. To address this problem, we propose an inverse reinforcement learning method that allows us to estimate the policy parameters of the learning agent which can then be used to estimate its reward function. Our method relies on a new variant of the behavior cloning algorithm, which we call bundle behavior cloning, and uses a small number of trajectories generated by the learning agent's policy at different points in time to learn a set of policies that match the distribution of actions observed in the sampled trajectories. We then use the cloned policies to train a neural network model that estimates the reward function of the learning agent. We provide a theoretical analysis to show a complexity result on bound guarantees for our method that beats standard behavior cloning as well as numerical experiments for a reinforcement learning problem that validate the proposed method.
format Preprint
id arxiv_https___arxiv_org_abs_2410_14135
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Inverse Reinforcement Learning from Non-Stationary Learning Agents
Sivakumar, Kavinayan P.
Shen, Yi
Bell, Zachary
Nivison, Scott
Chen, Boyuan
Zavlanos, Michael M.
Machine Learning
Artificial Intelligence
In this paper, we study an inverse reinforcement learning problem that involves learning the reward function of a learning agent using trajectory data collected while this agent is learning its optimal policy. To address this problem, we propose an inverse reinforcement learning method that allows us to estimate the policy parameters of the learning agent which can then be used to estimate its reward function. Our method relies on a new variant of the behavior cloning algorithm, which we call bundle behavior cloning, and uses a small number of trajectories generated by the learning agent's policy at different points in time to learn a set of policies that match the distribution of actions observed in the sampled trajectories. We then use the cloned policies to train a neural network model that estimates the reward function of the learning agent. We provide a theoretical analysis to show a complexity result on bound guarantees for our method that beats standard behavior cloning as well as numerical experiments for a reinforcement learning problem that validate the proposed method.
title Inverse Reinforcement Learning from Non-Stationary Learning Agents
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2410.14135