Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash 92 tok/s
Gemini 2.5 Pro 49 tok/s Pro
GPT-5 Medium 32 tok/s
GPT-5 High 40 tok/s Pro
GPT-4o 83 tok/s
GPT OSS 120B 467 tok/s Pro
Kimi K2 197 tok/s Pro
2000 character limit reached

Towards Bayesian Data Selection (2406.12560v2)

Published 18 Jun 2024 in stat.ML, cs.AI, cs.LG, math.ST, and stat.TH

Abstract: A wide range of machine learning algorithms iteratively add data to the training sample. Examples include semi-supervised learning, active learning, multi-armed bandits, and Bayesian optimization. We embed this kind of data addition into decision theory by framing data selection as a decision problem. This paves the way for finding Bayes-optimal selections of data. For the illustrative case of self-training in semi-supervised learning, we derive the respective Bayes criterion. We further show that deploying this criterion mitigates the issue of confirmation bias by empirically assessing our method for generalized linear models, semi-parametric generalized additive models, and Bayesian neural networks on simulated and real-world data.

List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

Summary

We haven't generated a summary for this paper yet.

Dice Question Streamline Icon: https://streamlinehq.com

Follow-up Questions

We haven't generated follow-up questions for this paper yet.

Authors (1)

Don't miss out on important new AI/ML research

See which papers are being discussed right now on X, Reddit, and more:

“Emergent Mind helps me see which AI papers have caught fire online.”

Philip

Philip

Creator, AI Explained on YouTube