paper-with-me

홈 › Papers

NECO: NEural Collapse Based Out-of-distribution detection

2023-10-10 · Mouïn Ben Ammar, Nacim Belkhir, Sebastian Popescu, Antoine Manzanera, Gianni Franchi

Detecting out-of-distribution (OOD) data is a critical challenge in machine learning due to model overconfidence, often without awareness of their epistemological limits. We hypothesize that `neural collapse'', a phenomenon affecting in-distribution data for models trained beyond loss convergence, also influences OOD data. To benefit from this interplay, we introduce NECO, a novel post-hoc method for OOD detection, which leverages the geometric properties of `neural collapse'' and of principal component spaces to identify OOD data. Our extensive experiments demonstrate that NECO achieves state-of-the-art results on both small and large-scale OOD detection tasks while exhibiting strong generalization capabilities across different network architectures. Furthermore, we provide a theoretical explanation for the effectiveness of our method in OOD detection. Code is available at https://gitlab.com/drti/neco

📄 PDF Abstract BibTeX arXiv:2310.06823

Code (1)

https://gitlab.com/drti/neco 공식 구현 pytorch

Tasks

Out-of-Distribution Detection

Similar Papers 제목 키워드 기반

Energy-Aware NECO for Single-Pass Pixel-wise Out-of-Distribution Detection in Semantic Segmentation

2026-05-28 · Boyuan Zhang, Huanshan Huang, Yifei Cao arxiv

Reliable semantic segmentation for mobile robots requires both accurate dense prediction and robust uncertainty estimation under distribution shift. Strong uncertainty baselines such as Monte Carlo Dropout often require …

Out-of-Distribution DetectionSemantic Segmentation

FineControlNet: Fine-level Text Control for Image Generation with Spatially Aligned Text Control Injection

2023-12-14 · Hongsuk Choi, Isaac Kasahara, Selim Engin, Moritz Graule 외

Recently introduced ControlNet has the ability to steer the text-driven image generation process with geometric input such as human 2D pose, or edge features. While ControlNet provides control over the geometric form of …

Image Generation

LineCounter: Learning Handwritten Text Line Segmentation by Counting

2021-05-24 · Deng Li, Yue Wu, Yicong Zhou

Handwritten Text Line Segmentation (HTLS) is a low-level but important task for many higher-level document processing tasks like handwritten text recognition. It is often formulated in terms of semantic segmentation or o…

Handwritten Text Recognitionobject-detectionObject DetectionSegmentation+1

LineConGraphs: Line Conversation Graphs for Effective Emotion Recognition using Graph Neural Networks

2023-12-04 · Gokul S Krishnan, Sarala Padi, Craig S. Greenberg, Balaraman Ravindran 외

Emotion Recognition in Conversations (ERC) is a critical aspect of affective computing, and it has many practical applications in healthcare, education, chatbots, and social media platforms. Earlier approaches for ERC an…

Emotion RecognitionGraph Attentiongraph constructionGraph Neural Network

LaneCorrect: Self-supervised Lane Detection

2024-04-23 · Ming Nie, Xinyue Cai, Hang Xu, Li Zhang

Lane detection has evolved highly functional autonomous driving system to understand driving scenes even under complex environments. In this paper, we work towards developing a generalized computer vision system able to …

Autonomous DrivingLane Detection