REVIEW 2 cited by
Combining Offline Causal Inference and Online Bandit Learning for Data Driven Decision
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
A fundamental question for companies with large amount of logged data is: How to use such logged data together with incoming streaming data to make good decisions? Many companies currently make decisions via online A/B tests, but wrong decisions during testing hurt users' experiences and cause irreversible damage. A typical alternative is offline causal inference, which analyzes logged data alone to make decisions. However, these decisions are not adaptive to the new incoming data, and so a wrong decision will continuously hurt users' experiences. To overcome the aforementioned limitations, we propose a framework to unify offline causal inference algorithms (e.g., weighting, matching) and online learning algorithms (e.g., UCB, LinUCB). We propose novel algorithms and derive bounds on the decision accuracy via the notion of "regret". We derive the first upper regret bound for forest-based online bandit algorithms. Experiments on two real datasets show that our algorithms outperform other algorithms that use only logged data or online feedbacks, or algorithms that do not use the data properly.
Forward citations
Cited by 2 Pith papers
-
Contextual Online Pricing with (Biased) Offline Data
Contextual online pricing can safely incorporate biased offline data, achieving the standard square-root-of-T worst-case regret and better rates when the offline bias is small.
-
Best Arm Identification with Possibly Biased Offline Data
LUCB-H adaptively combines offline and online data for best arm identification, matching or beating standard LUCB depending on whether the historical data is helpful or misleading.
Discussion (0). Continue with ORCID to comment.