Saved in:
Bibliographic Details
Main Author: Zhao, Xingyu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.14833
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910864830889984
author Zhao, Xingyu
author_facet Zhao, Xingyu
contents Deep learning (DL) has demonstrated significant potential across various safety-critical applications, yet ensuring its robustness remains a key challenge. While adversarial robustness has been extensively studied in worst-case scenarios, probabilistic robustness (PR) offers a more practical perspective by quantifying the likelihood of failures under stochastic perturbations. This paper provides a concise yet comprehensive overview of PR, covering its formal definitions, evaluation and enhancement methods. We introduce a reformulated ''min-max'' optimisation framework for adversarial training specifically designed to improve PR. Furthermore, we explore the integration of PR verification evidence into system-level safety assurance, addressing challenges in translating DL model-level robustness to system-level claims. Finally, we highlight open research questions, including benchmarking PR evaluation methods, extending PR to generative AI tasks, and developing rigorous methodologies and case studies for system-level integration.
format Preprint
id arxiv_https___arxiv_org_abs_2502_14833
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Probabilistic Robustness in Deep Learning: A Concise yet Comprehensive Guide
Zhao, Xingyu
Machine Learning
Deep learning (DL) has demonstrated significant potential across various safety-critical applications, yet ensuring its robustness remains a key challenge. While adversarial robustness has been extensively studied in worst-case scenarios, probabilistic robustness (PR) offers a more practical perspective by quantifying the likelihood of failures under stochastic perturbations. This paper provides a concise yet comprehensive overview of PR, covering its formal definitions, evaluation and enhancement methods. We introduce a reformulated ''min-max'' optimisation framework for adversarial training specifically designed to improve PR. Furthermore, we explore the integration of PR verification evidence into system-level safety assurance, addressing challenges in translating DL model-level robustness to system-level claims. Finally, we highlight open research questions, including benchmarking PR evaluation methods, extending PR to generative AI tasks, and developing rigorous methodologies and case studies for system-level integration.
title Probabilistic Robustness in Deep Learning: A Concise yet Comprehensive Guide
topic Machine Learning
url https://arxiv.org/abs/2502.14833