TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Deprecating A/B tests with offline policy evaluation

1 pointsby econtiabout 4 years ago
I recently read about a technique in statistics&#x2F;ML called &quot;offline policy evaluation&quot;.<p>The idea is that you can evaluate how new policies will perform by using historical data generated under a previous policy. For example, rather than testing a new fraud policy in an A&#x2F;B test, you can use historical logs to determine if the new policy will outperform the existing one. This seems like it could be a great step before A&#x2F;B testing new policies.<p>I whipped up some example code to test out what would be considered the &quot;hello world&quot; of offline policy evaluation if anyone is curious: https:&#x2F;&#x2F;github.com&#x2F;banditml&#x2F;offline-policy-evaluation&#x2F;blob&#x2F;main&#x2F;examples&#x2F;Inverse%20propensity%20scoring.ipynb<p>My question to you is -- have any of you have tried this or do any of your currently use OPE at your companies?

no comments

no comments