Black-box Certification and Learning under Adversarial Perturbations

Authors: Hassan Ashtiani, Vinayak Pathak, Ruth Urner

ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We formally study the problem of classification under adversarial perturbations from a learner s perspective as well as a third-party who aims at certifying the robustness of a given black-box classifier. We analyze a PAC-type framework of semisupervised learning and identify possibility and impossibility results for proper learning of VCclasses in this setting. We further introduce a new setting of black-box certification under limited query budget, and analyze this for various classes of predictors and perturbation.
Researcher Affiliation Collaboration 1Department of Computing and Software, Mc Master University, Hamilton, ON, Canada 2Scotiabank, Toronto, ON, Canada 3Lassonde School of Engineering, EECS Department, York University, Toronto, ON, Canada.
Pseudocode No The paper contains formal definitions, theorems, and proof sketches, but no sections or figures labeled 'Pseudocode' or 'Algorithm', nor any structured, code-like steps for a procedure.
Open Source Code No The paper does not provide any statement about making its source code publicly available, nor does it include links to a code repository.
Open Datasets No This is a theoretical paper and does not describe the use of any datasets for training or experimentation.
Dataset Splits No This paper is theoretical and does not describe experimental validation using dataset splits.
Hardware Specification No The paper focuses on theoretical analysis and does not describe any experimental setup involving specific hardware specifications.
Software Dependencies No The paper focuses on theoretical analysis and does not describe any specific software dependencies or versions used for implementation or experimentation.
Experiment Setup No The paper focuses on theoretical analysis and does not describe any experimental setup with specific hyperparameters, training configurations, or system-level settings.