Blockers
rebels in the deep state
Yaoliang Yu
Vector Institute
About Me
Yaoliang Yu (于耀亮) is currently a faculty in the David R. Cheriton School of Computer Science at University of Waterloo, as well as at the Vector institute. His main research interests include generative models, convex and nonconvex optimization, trustworthy machine learning, and applications in computer vision and natural language processing.
Some Publications
(2026).
Provably Adaptive Linear Approximation for the Shapley Value and Beyond.
International Conference on Machine Learning (ICML).
(2025).
Stochastic Forward-Backward Deconvolution: Training Diffusion Models with Finite Noisy Datasets.
International Conference on Machine Learning (ICML).
(2025).
Leveraging Variable Sparsity to Refine Pareto Stationarity in Multi-Objective Optimization.
International Conference on Learning Representations (ICLR).
(2024).
Faster Approximation of Probabilistic and Distributional Values via Least Squares.
International Conference on Learning Representations (ICLR).
(2023).
Exploring the Limits of Model-Targeted Indiscriminate Data Poisoning Attacks.
International Conference on Machine Learning (ICML).
(2021).
Demystifying and Generalizing BinaryConnect.
Advances in Neural Information Processing Systems (NeurIPS).
(2019).
Multivariate Triangular Quantile Maps for Novelty Detection.
Advances in Neural Information Processing Systems (NeurIPS).
(2019).
Sum-of-squares Polynomial Flow.
International Conference on Machine Learning (ICML).
(2018).
Deep Homogeneous Mixture Models: Representation, Separation and Approximation.
Advances in Neural Information Processing Systems (NeurIPS).
(2017).
Generalized Conditional Gradient for Sparse Estimation.
Journal of Machine Learning Research (JMLR).
(2015).
Minimizing Nonconvex Non-Separable Functions.
International Conference on Artificial Intelligence and Statistics (AISTATS).
(2013).
Characterizing the Representer Theorem.
International Conference on Machine Learning (ICML).
(2013).
On Decomposing the Proximal Map.
Advances in Neural Information Processing Systems (NeurIPS).
(2011).
Rank/Norm Regularization with Closed-Form Solutions: Application to Subspace Clustering.
Conference on Uncertainty in Artificial Intelligence (UAI).
(2009).
A General Projection Property for Distribution Families.
Advances in Neural Information Processing Systems (NeurIPS).
Recent Courses
Recent Readings
Bad Blood
secrets and lies in a silicon valley startup
The American Way of Killing
the invention of an epidemic
The Map of My Life
人生的地图