Sequential Counterfactual Decision-Making Under Confounded Reward
Erik Skalnes
TL;DR
This paper explores the limitations of random trials in decision-making when confounding factors affect the cause-effect relationship, using a formal counterfactual framework with soft interventions.
Contribution
It introduces a formal counterfactual policy-space model that incorporates the agent's natural preferences through soft interventions to analyze confounded decision-making.
Findings
Identifies limitations of random trials under confounding.
Proposes a counterfactual policy-space framework.
Provides insights into decision-making with confounded rewards.
Abstract
We investigate the limitations of random trials when the cause of interest is confounded with the effect by formalizing a counterfactual policy-space where the agent's natural predilection is input to a soft-intervention.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGame Theory and Applications · Economic Policies and Impacts · Decision-Making and Behavioral Economics