Saved in:
Bibliographic Details
Main Authors: Guglielmo, Gianluca, Masana, Marc
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.12849
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909500009611264
author Guglielmo, Gianluca
Masana, Marc
author_facet Guglielmo, Gianluca
Masana, Marc
contents In real-world applications, machine learning models must reliably detect Out-of-Distribution (OoD) samples to prevent unsafe decisions. Current OoD detection methods often rely on analyzing the logits or the embeddings of the penultimate layer of a neural network. However, little work has been conducted on the exploitation of the rich information encoded in intermediate layers. To address this, we analyze the discriminative power of intermediate layers and show that they can positively be used for OoD detection. Therefore, we propose to regularize intermediate layers with an energy-based contrastive loss, and by grouping multiple layers in a single aggregated response. We demonstrate that intermediate layer activations improves OoD detection performance by running a comprehensive evaluation across multiple datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2502_12849
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Leveraging Intermediate Representations for Better Out-of-Distribution Detection
Guglielmo, Gianluca
Masana, Marc
Machine Learning
Computer Vision and Pattern Recognition
I.4.9
In real-world applications, machine learning models must reliably detect Out-of-Distribution (OoD) samples to prevent unsafe decisions. Current OoD detection methods often rely on analyzing the logits or the embeddings of the penultimate layer of a neural network. However, little work has been conducted on the exploitation of the rich information encoded in intermediate layers. To address this, we analyze the discriminative power of intermediate layers and show that they can positively be used for OoD detection. Therefore, we propose to regularize intermediate layers with an energy-based contrastive loss, and by grouping multiple layers in a single aggregated response. We demonstrate that intermediate layer activations improves OoD detection performance by running a comprehensive evaluation across multiple datasets.
title Leveraging Intermediate Representations for Better Out-of-Distribution Detection
topic Machine Learning
Computer Vision and Pattern Recognition
I.4.9
url https://arxiv.org/abs/2502.12849