We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Reinforcement Learning Beyond Expectation

Abstract: The inputs and preferences of human users are important considerations in situations where these users interact with autonomous cyber or cyber-physical systems. In these scenarios, one is often interested in aligning behaviors of the system with the preferences of one or more human users. Cumulative prospect theory (CPT) is a paradigm that has been empirically shown to model a tendency of humans to view gains and losses differently. In this paper, we consider a setting where an autonomous agent has to learn behaviors in an unknown environment. In traditional reinforcement learning, these behaviors are learned through repeated interactions with the environment by optimizing an expected utility. In order to endow the agent with the ability to closely mimic the behavior of human users, we optimize a CPT-based cost. We introduce the notion of the CPT-value of an action taken in a state, and establish the convergence of an iterative dynamic programming-based approach to estimate this quantity. We develop two algorithms to enable agents to learn policies to optimize the CPT-vale, and evaluate these algorithms in environments where a target state has to be reached while avoiding obstacles. We demonstrate that behaviors of the agent learned using these algorithms are better aligned with that of a human user who might be placed in the same environment, and is significantly improved over a baseline that optimizes an expected utility.
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Systems and Control (eess.SY); Optimization and Control (math.OC)
Cite as: arXiv:2104.00540 [cs.LG]
  (or arXiv:2104.00540v1 [cs.LG] for this version)

Submission history

From: Bhaskar Ramasubramanian [view email]
[v1] Mon, 29 Mar 2021 20:35:25 GMT (263kb,D)

Link back to: arXiv, form interface, contact.