Terra Cognita

Efficiently Breaking the Curse of Horizon in Off-Policy Evaluation with Double Reinforcement Learning

arxiv-nonexclusive link only — licence forbids redistribution arxiv 2019

Nathan Kallus, Masatoshi Uehara

arXiv preprint (stat.ML, cs.LG, math.OC).

Source ↗

ABSTRACT · 0.0 MB · sha256 e1d60426428c…

Concepts this teaches

Not yet mapped to any concept.

A resource only becomes useful here once a curator has anchored it to concepts at specific pages. Until then it is a book on a shelf.