Y
Yizhen Wang
Researcher at University of California, San Diego
Publications - 20
Citations - 406
Yizhen Wang is an academic researcher from University of California, San Diego. The author has contributed to research in topics: Robustness (computer science) & Computer science. The author has an hindex of 7, co-authored 12 publications receiving 295 citations.
Papers
More filters
Proceedings ArticleDOI
Pufferfish Privacy Mechanisms for Correlated Data
TL;DR: This work provides the first mechanism -- the Wasserstein Mechanism -- which applies to any general Pufferfish framework, and provides an additional mechanism that applies to some practical cases such as physical activity measurements across time, and is computationally efficient.
Posted Content
Data Poisoning Attacks against Online Learning
Yizhen Wang,Kamalika Chaudhuri +1 more
TL;DR: A systematic investigation of data poisoning attacks for online learning is initiated, and a general attack strategy is proposed, formulated as an optimization problem, that applies to both settings with some modifications.
Posted Content
Analyzing the Robustness of Nearest Neighbors to Adversarial Examples
TL;DR: This work introduces a theoretical framework analogous to bias-variance theory for understanding why adversarial examples arise, and uses this framework to analyze the robustness of a canonical non-parametric classifier - the k-nearest neighbors.
Proceedings Article
Analyzing the Robustness of Nearest Neighbors to Adversarial Examples
TL;DR: In this article, the robustness of a canonical non-parametric classifier, the k-nearest neighbors, was analyzed and it was shown that its robustness properties depend critically on the value of k.
Proceedings Article
Robustness for Non-Parametric Classification: A Generic Attack and Defense
TL;DR: In this paper, the authors take a holistic look at adversarial examples for non-parametric classifiers, including nearest neighbors, decision trees, and random forests, and derive an optimally robust classifier, which is analogous to the Bayes Optimal.