paper-with-me

홈 › Papers

Don't Collapse Your Features: Why CenterLoss Hurts OOD Detection and Multi-Scale Mahalanobis Wins

2026-04-10 · Rahul D Ray arxiv

The ability to detect out-of-distribution (OOD) inputs is fundamental to safe deployment of machine learning systems. Yet, current methods often rely on feature representations that are optimised solely for classification accuracy, neglecting the distinct requirements of epistemic uncertainty. We introduce GOEN (Geometry-Optimised Epistemic Network), a simple pipeline that combines multi-scale features, L2 normalisation, Mahalanobis distance, and a calibration head trained with real hard OOD examples. Through systematic ablation we uncover a counter-intuitive finding: CenterLoss, a popular regulariser for feature compactness, significantly degrades OOD detection performance, reducing average OOD AUROC from 0.9483 to 0.9366 despite improving classification accuracy. The best variant, GOEN-NoCenterLoss, achieves an average OOD AUROC of 0.9483, surpassing all baselines including deep ensembles (0.8827), KNN (0.8967), and ODIN (0.8870) on CIFAR-10 benchmarks, while maintaining competitive in-distribution accuracy. Our results challenge the prevailing assumption that better classification geometry automatically leads to better epistemic uncertainty. Instead, we show that overly tight feature clusters compress inter-class margins and distort the covariance structure needed for effective OOD detection. GOEN is efficient, training in under 20 minutes on a single GPU, and provides a practical blueprint for building AI systems that reliably recognise their own limitations.

📄 PDF Abstract BibTeX arXiv:2605.21493

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Multi-modal Deep Learning

2024-03-06 · Chen Yuhua

This article investigates deep learning methodologies for single-modality clinical data analysis, as a crucial precursor to multi-modal medical research. Building on Guo JingYuan's work, the study refines clinical data p…

Deep LearningTime SeriesTransfer Learning

Modulate Your Spectrum in Self-Supervised Learning

2023-05-26 · Xi Weng, Yunhao Ni, Tengwei Song, Jie Luo 외

Whitening loss offers a theoretical guarantee against feature collapse in self-supervised learning (SSL) with joint embedding architectures. Typically, it involves a hard whitening approach, transforming the embedding an…

object-detectionObject DetectionSelf-Supervised Learning

Run Away From your Teacher: Understanding BYOL by a Novel Self-Supervised Approach

2020-11-22 · Haizhou Shi, Dongliang Luo, Siliang Tang, Jian Wang 외

Recently, a newly proposed self-supervised framework Bootstrap Your Own Latent (BYOL) seriously challenges the necessity of negative samples in contrastive learning frameworks. BYOL works like a charm despite the fact th…

Contrastive LearningSelf-Supervised Learning

Run Away From your Teacher: a New Self-Supervised Approach Solving the Puzzle of BYOL

2021-01-01 · Haizhou Shi, Dongliang Luo, Siliang Tang, Jian Wang 외

Recently, a newly proposed self-supervised framework Bootstrap Your Own Latent (BYOL) seriously challenges the necessity of negative samples in contrastive-based learning frameworks. BYOL works like a charm despite the f…

Self-Supervised Learning

You Are Your Own Best Teacher: Achieving Centralized-level Performance in Federated Learning under Heterogeneous and Long-tailed Data

2025-03-10 · Shanshan Yan, Zexi Li, Chao Wu, Meng Pang 외

Data heterogeneity, stemming from local non-IID data and global long-tailed distributions, is a major challenge in federated learning (FL), leading to significant performance gaps compared to centralized learning. Previo…

Federated LearningRepresentation Learning