Avatar

Kaizhao Liang

Champaign,United State.

About

I am a undergraduate affliated with Secure learning Lab in University of Illinois, Urbana-Champaign.

Currently, I have the honor and privilege to work with Professor Bo Li in Department of Computer Science in UIUC.

Interests

I have focused on studying the robustness and security of models in various domains, such as computer vision and natural language processing.

Recently, I started to investigate the sparsification of neural networks and the intriguing properties of the glorious Stochastic Gradient Descent. For future, I am interested in bridging the gap between theory and practitioning.

Publications

Big but Imperceptible Adversarial Perturbations via Semantic Manipulation [CVPR 2019 advML workshop] Anand Bhattad, Min Jin Chong, Kaizhao Liang, Bo Li, David A. Forsyth

Relevant Projects

Semantic incosistency in deepfake video

Look for a way to detect manipulated videos

Interval bound propagation pytorch

Exploring to train certified defense against adversarial attacks with limited resources by "tighter" objective

Learning to Run

TRPO and DDPG implementation to teach robot to walk in simulations.

Fake News Challenge

Detecting sentimental/semantic inconsistency in fake news.

Contact

kl2@illinois.edu