FAST-Q estimates counterfactual recommendation outcomes offline by learning policy-invariant state representations, and reports business metric gains on a gaming platform.
Hasegawa- Johnson, and Thomas S
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FAST-Q: Fast-track Exploration with Adversarially Balanced State Representations for Counterfactual Action Estimation in Offline Reinforcement Learning
FAST-Q estimates counterfactual recommendation outcomes offline by learning policy-invariant state representations, and reports business metric gains on a gaming platform.