Saved in:
Bibliographic Details
Main Authors: Vishwakarma, Harit, Reid, Chen, Tay, Sui Jiet, Namburi, Satya Sai Srinath, Sala, Frederic, Vinayak, Ramya Korlakai
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.16188
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913328799940608
author Vishwakarma, Harit
Reid
Chen
Tay, Sui Jiet
Namburi, Satya Sai Srinath
Sala, Frederic
Vinayak, Ramya Korlakai
author_facet Vishwakarma, Harit
Reid
Chen
Tay, Sui Jiet
Namburi, Satya Sai Srinath
Sala, Frederic
Vinayak, Ramya Korlakai
contents Auto-labeling is an important family of techniques that produce labeled training sets with minimum manual labeling. A prominent variant, threshold-based auto-labeling (TBAL), works by finding a threshold on a model's confidence scores above which it can accurately label unlabeled data points. However, many models are known to produce overconfident scores, leading to poor TBAL performance. While a natural idea is to apply off-the-shelf calibration methods to alleviate the overconfidence issue, such methods still fall short. Rather than experimenting with ad-hoc choices of confidence functions, we propose a framework for studying the \emph{optimal} TBAL confidence function. We develop a tractable version of the framework to obtain \texttt{Colander} (Confidence functions for Efficient and Reliable Auto-labeling), a new post-hoc method specifically designed to maximize performance in TBAL systems. We perform an extensive empirical evaluation of our method \texttt{Colander} and compare it against methods designed for calibration. \texttt{Colander} achieves up to 60\% improvements on coverage over the baselines while maintaining auto-labeling error below $5\%$ and using the same amount of labeled data as the baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2404_16188
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Pearls from Pebbles: Improved Confidence Functions for Auto-labeling
Vishwakarma, Harit
Reid
Chen
Tay, Sui Jiet
Namburi, Satya Sai Srinath
Sala, Frederic
Vinayak, Ramya Korlakai
Machine Learning
Artificial Intelligence
Auto-labeling is an important family of techniques that produce labeled training sets with minimum manual labeling. A prominent variant, threshold-based auto-labeling (TBAL), works by finding a threshold on a model's confidence scores above which it can accurately label unlabeled data points. However, many models are known to produce overconfident scores, leading to poor TBAL performance. While a natural idea is to apply off-the-shelf calibration methods to alleviate the overconfidence issue, such methods still fall short. Rather than experimenting with ad-hoc choices of confidence functions, we propose a framework for studying the \emph{optimal} TBAL confidence function. We develop a tractable version of the framework to obtain \texttt{Colander} (Confidence functions for Efficient and Reliable Auto-labeling), a new post-hoc method specifically designed to maximize performance in TBAL systems. We perform an extensive empirical evaluation of our method \texttt{Colander} and compare it against methods designed for calibration. \texttt{Colander} achieves up to 60\% improvements on coverage over the baselines while maintaining auto-labeling error below $5\%$ and using the same amount of labeled data as the baselines.
title Pearls from Pebbles: Improved Confidence Functions for Auto-labeling
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2404.16188