Learning Outlier-Aware Representation with Synthetic Boundary Samples

Jen Tzung Chien*, Kuan Chen

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Out-of-distribution (OOD) detection provides an essential scheme to build a reliable deep learning system. Some of the previous works additionally acquired OOD samples for model training or hyperparameter tuning. However, OOD samples are usually missing in real-world scenarios and the prior knowledge of these samples is commonly unavailable. To cope with these issues, this paper presents a novel approach to synthesizing the informative samples near the boundary between in-distribution and OOD which sufficiently reflect OOD samples in training stage. These synthetic boundary samples are located right outside in-distribution (ID) where the latent variable is distributed by a multivariate Gaussian. Accordingly, this paper presents the outlier-aware representation learning which utilizes the synthetic boundary samples to train an OOD detector by only using the unlabeled ID data. The model is then trained to learn a compact decision boundary between ID and OOD samples. The experiments demonstrate that the proposed methods outperform state-of-the-art performance in presence of different OOD data.

Original languageEnglish
Title of host publicationProceedings of the 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing, MLSP 2023
EditorsDanilo Comminiello, Michele Scarpiniti
PublisherIEEE Computer Society
ISBN (Electronic)9798350324112
DOIs
StatePublished - 2023
Event33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023 - Rome, Italy
Duration: 17 Sep 202320 Sep 2023

Publication series

NameIEEE International Workshop on Machine Learning for Signal Processing, MLSP
Volume2023-September
ISSN (Print)2161-0363
ISSN (Electronic)2161-0371

Conference

Conference33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023
Country/TerritoryItaly
CityRome
Period17/09/2320/09/23

Keywords

  • Out-of-distribution detection
  • computer vision
  • contrastive learning
  • natural language processing

Fingerprint

Dive into the research topics of 'Learning Outlier-Aware Representation with Synthetic Boundary Samples'. Together they form a unique fingerprint.

Cite this