paper-with-me

홈 › Papers

It's an Alignment, Not a Trade-off: Revisiting Bias and Variance in Deep Models

2023-10-13 · Lin Chen, Michal Lukasik, Wittawat Jitkrittum, Chong You, Sanjiv Kumar

Classical wisdom in machine learning holds that the generalization error can be decomposed into bias and variance, and these two terms exhibit a \emph{trade-off}. However, in this paper, we show that for an ensemble of deep learning based classification models, bias and variance are \emph{aligned} at a sample level, where squared bias is approximately \emph{equal} to variance for correctly classified sample points. We present empirical evidence confirming this phenomenon in a variety of deep learning models and datasets. Moreover, we study this phenomenon from two theoretical perspectives: calibration and neural collapse. We first show theoretically that under the assumption that the models are well calibrated, we can observe the bias-variance alignment. Second, starting from the picture provided by the neural collapse theory, we show an approximate correlation between bias and variance.

📄 PDF Abstract BibTeX arXiv:2310.09250

Code (0)

등록된 구현이 없습니다.

Tasks

Deep Learning

Similar Papers 제목 키워드 기반

Rethinking Soft Labels for Knowledge Distillation: A Bias-Variance Tradeoff Perspective

2021-02-01 · Helong Zhou, Liangchen Song, Jiajie Chen, Ye Zhou 외

Knowledge distillation is an effective approach to leverage a well-trained network or an ensemble of them, named as the teacher, to guide the training of a student network. The outputs from the teacher network are used a…

Knowledge Distillation

The Unbearable Weight of Massive Privilege: Revisiting Bias-Variance Trade-Offs in the Context of Fair Prediction

2023-02-17 · Falaah Arif Khan, Julia Stoyanovich

In this paper we revisit the bias-variance decomposition of model error from the perspective of designing a fair classifier: we are motivated by the widely held socio-technical belief that noise variance in large dataset…

Fairness

From Noisy Traces to Stable Gradients: Bias-Variance Optimized Preference Optimization for Aligning Large Reasoning Models

2025-10-06 · Mingkang Zhu, Xi Chen, Bei Yu, Hengshuang Zhao 외 arxiv

Large reasoning models (LRMs) generate intermediate reasoning traces before producing final answers, yielding strong gains on multi-step and mathematical tasks. Yet aligning LRMs with human preferences, a crucial prerequ…

On the Bias-Variance Tradeoff: Textbooks Need an Update

2019-12-17 · Brady Neal

The main goal of this thesis is to point out that the bias-variance tradeoff is not always true (e.g. in neural networks). We advocate for this lack of universality to be acknowledged in textbooks and taught in introduct…

Deep Reinforcement LearningReinforcement Learning

On lower bounds for the bias-variance trade-off

2020-05-30 · Alexis Derumigny, Johannes Schmidt-Hieber

It is a common phenomenon that for high-dimensional and nonparametric statistical models, rate-optimal estimators balance squared bias and variance. Although this balancing is widely observed, little is known whether met…