Ji Gao
Ji Gao
PhD Candidate, Computer Science Department, Univerisity of Virginia
Bestätigte E-Mail-Adresse bei - Startseite
Zitiert von
Zitiert von
Black-box generation of adversarial text sequences to evade deep learning classifiers
J Gao, J Lanchantin, ML Soffa, Y Qi
2018 IEEE Security and Privacy Workshops (SPW), 50-56, 2018
Deepcloak: Masking deep neural network models for robustness against adversarial samples
J Gao, B Wang, Z Lin, W Xu, Y Qi
arXiv preprint arXiv:1702.06763, 2017
A theoretical framework for robustness of (deep) classifiers against adversarial examples
B Wang, J Gao, Y Qi
arXiv preprint arXiv:1612.00334, 2016
STLnet: Signal Temporal Logic Enforced Multivariate Recurrent Neural Networks.
M Ma, J Gao, L Feng, JA Stankovic
2020 Conference on Neural Information Processing Systems (Neurips 2020), 2020
Learning and certification under instance-targeted poisoning
J Gao, A Karbasi, M Mahmoody
Uncertainty in Artificial Intelligence, 2135-2145, 2021
Unsupervised graph alignment with wasserstein distance discriminator
J Gao, X Huang, J Li
Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data …, 2021
A fast and scalable joint estimator for learning multiple related sparse gaussian graphical models
B Wang, J Gao, Y Qi
Artificial Intelligence and Statistics, 1168-1177, 2017
Deletion Inference, Reconstruction, and Compliance in Machine (Un) Learning
J Gao, S Garg, M Mahmoody, PN Vasudevan
arXiv preprint arXiv:2202.03460, 2022
Exploring the naturalness of buggy code with recurrent neural networks
J Lanchantin, J Gao
arXiv preprint arXiv:1803.08793, 2018
Polynomial-time targeted attacks on coin tossing for any number of corruptions
O Etesami, J Gao, S Mahloujifar, M Mahmoody
Theory of Cryptography Conference, 718-750, 2021
Pragmatic Evaluation of Adversarial Examples in Natural Language
J Morris, E Lifland, J Gao, J Lanchantin, Y Ji, Y Qi
MCTSBug: Generating Adversarial Text Sequences via Monte Carlo Tree Search and Homoglyph Attack
J Gao, J Lanchantin, Y Qi
Das System kann den Vorgang jetzt nicht ausführen. Versuchen Sie es später erneut.
Artikel 1–12