Learning Sample Difficulty from Pre-trained Models for Reliable Prediction

Published: 21 Sept 2023, Last Modified: 02 Nov 2023NeurIPS 2023 posterEveryoneRevisionsBibTeX
Keywords: uncertainty calibration, sample difficulty, reliable prediction
TL;DR: We propose to utilize large-scale pre-trained models to guide downstream model training with sample difficulty-aware entropy regularization to improve accuracy and uncertainty calibration
Abstract: Large-scale pre-trained models have achieved remarkable success in many applications, but how to leverage them to improve the prediction reliability of downstream models is undesirably under-explored. Moreover, modern neural networks have been found to be poorly calibrated and make overconfident predictions regardless of inherent sample difficulty and data uncertainty. To address this issue, we propose to utilize large-scale pre-trained models to guide downstream model training with sample difficulty-aware entropy regularization. Pre-trained models that have been exposed to large-scale datasets and do not overfit the downstream training classes enable us to measure each training sample’s difficulty via feature-space Gaussian modeling and relative Mahalanobis distance computation. Importantly, by adaptively penalizing overconfident prediction based on the sample difficulty, we simultaneously improve accuracy and uncertainty calibration across challenging benchmarks (e.g., +0.55% ACC and −3.7% ECE on ImageNet1k using ResNet34), consistently surpassing competitive baselines for reliable prediction. The improved uncertainty estimate further improves selective classification (abstaining from erroneous predictions) and out-of-distribution detection.
Supplementary Material: zip
Submission Number: 11276
Loading