hongyanz
Assistant Professor of CS at UWaterloo and Vector Institute. Leader of SafeAI Lab. Previously Ph.D. at CMU MLD.
UWaterloo & Vector InstituteWaterloo, Canada
Pinned Repositories
deep-image-prior
Image restoration with neural networks but without learning.
Life-Long-Matrix-Completion
MATLAB code for NIPS 2016 paper "Noise-Tolerant Life-Long Matrix Completion via Adaptive Sampling"
multibranch
Codes for the paper "Deep Neural Networks with Multi-Branch Architectures Are Less Non-Convex"
PyTorch-GAN
PyTorch implementations of Generative Adversarial Networks.
Stackelberg-GAN
Codes for Stackelberg GAN
TRADES-smoothing
[JMLR] TRADES + random smoothing for certifiable robustness
Equivariant-Encryption-for-AI
Trade costly on-prem AI infrastructure for one blind API.
EAGLE
Official Implementation of EAGLE-1 (ICML'24) and EAGLE-2 (EMNLP'24)
RAIN
[ICLR'24] RAIN: Your Language Models Can Align Themselves without Finetuning
TRADES
TRADES (TRadeoff-inspired Adversarial DEfense via Surrogate-loss minimization)
hongyanz's Repositories
hongyanz/multibranch
Codes for the paper "Deep Neural Networks with Multi-Branch Architectures Are Less Non-Convex"
hongyanz/TRADES-smoothing
[JMLR] TRADES + random smoothing for certifiable robustness
hongyanz/Stackelberg-GAN
Codes for Stackelberg GAN
hongyanz/deep-image-prior
Image restoration with neural networks but without learning.
hongyanz/Life-Long-Matrix-Completion
MATLAB code for NIPS 2016 paper "Noise-Tolerant Life-Long Matrix Completion via Adaptive Sampling"
hongyanz/PyTorch-GAN
PyTorch implementations of Generative Adversarial Networks.