Online Reinforcement Learning in Non-Stationary Context-Driven Environments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hamadanian, Pouya, Nasr-Esfahany, Arash, Schwarzkopf, Malte, Sen, Siddartha, Alizadeh, Mohammad
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913767141408768
author Hamadanian, Pouya
Nasr-Esfahany, Arash
Schwarzkopf, Malte
Sen, Siddartha
Alizadeh, Mohammad
author_facet Hamadanian, Pouya
Nasr-Esfahany, Arash
Schwarzkopf, Malte
Sen, Siddartha
Alizadeh, Mohammad
contents We study online reinforcement learning (RL) in non-stationary environments, where a time-varying exogenous context process affects the environment dynamics. Online RL is challenging in such environments due to "catastrophic forgetting" (CF). The agent tends to forget prior knowledge as it trains on new experiences. Prior approaches to mitigate this issue assume task labels (which are often not available in practice), employ brittle regularization heuristics, or use off-policy methods that suffer from instability and poor performance. We present Locally Constrained Policy Optimization (LCPO), an online RL approach that combats CF by anchoring policy outputs on old experiences while optimizing the return on current experiences. To perform this anchoring, LCPO locally constrains policy optimization using samples from experiences that lie outside of the current context distribution. We evaluate LCPO in Mujoco, classic control and computer systems environments with a variety of synthetic and real context traces, and find that it outperforms a variety of baselines in the non-stationary setting, while achieving results on-par with a "prescient" agent trained offline across all context traces. LCPO's source code is available at https://github.com/pouyahmdn/LCPO.
format Preprint
id arxiv_https___arxiv_org_abs_2302_02182
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Online Reinforcement Learning in Non-Stationary Context-Driven Environments
Hamadanian, Pouya
Nasr-Esfahany, Arash
Schwarzkopf, Malte
Sen, Siddartha
Alizadeh, Mohammad
Machine Learning
Artificial Intelligence
We study online reinforcement learning (RL) in non-stationary environments, where a time-varying exogenous context process affects the environment dynamics. Online RL is challenging in such environments due to "catastrophic forgetting" (CF). The agent tends to forget prior knowledge as it trains on new experiences. Prior approaches to mitigate this issue assume task labels (which are often not available in practice), employ brittle regularization heuristics, or use off-policy methods that suffer from instability and poor performance. We present Locally Constrained Policy Optimization (LCPO), an online RL approach that combats CF by anchoring policy outputs on old experiences while optimizing the return on current experiences. To perform this anchoring, LCPO locally constrains policy optimization using samples from experiences that lie outside of the current context distribution. We evaluate LCPO in Mujoco, classic control and computer systems environments with a variety of synthetic and real context traces, and find that it outperforms a variety of baselines in the non-stationary setting, while achieving results on-par with a "prescient" agent trained offline across all context traces. LCPO's source code is available at https://github.com/pouyahmdn/LCPO.
title Online Reinforcement Learning in Non-Stationary Context-Driven Environments
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2302.02182