Shuang Liang

News 

• Jun 26: Invited talk at the SIAM OP26 mini-symposium "Global Optimization for Neural Networks".  

• Mar 26: Invited talk at the MFO workshop "Modern and Emerging Phenomena in Machine Learning". 

• Jan 26: Paper on chaos and fractals in gradient descent optimization accepted to ICLR 2026. 

• Nov 25: Awarded the John Fellowship (UCLA). 

• Jun 25: Invited talk at the Scalable Statistical Machine Learning Lab, UCSD. 

• Jan 25: Paper on implicit bias of mirror descent in ReLU networks accepted to ICLR 2025. 

Research 

I aim to better understand the learning of neural networks from optimization perspectives. In particular, I am excited about: 

• Optimization dynamics, e.g., training trajectories in parameter space; 

• Implicit bias of optimization algorithms, i.e., what types of functions an algorithm tends to select among (possibly) countless candidates that perform well on training data; 

• The influence of network architecture, optimizer, parameter initialization, step size, etc. 

Publications 

• Gradient Descent with Large Step Sizes: Chaos and Fractal Convergence Region
  Shuang Liang, Guido Montúfar  
  ICLR 2026
  [arXiv] [OpenReview]

• Implicit Bias of Mirror Flow for Shallow Neural Networks in Univariate Regression
  Shuang Liang, Guido Montúfar
  ICLR 2025 (Spotlight)
  [arXiv] [OpenReview] [Video]

• Pull-back Geometry of Persistent Homology Encodings
  Shuang Liang, Renata Turkeš, Jiayi Li, Nina Otter, Guido Montúfar
  TMLR 2024
  [arXiv] [OpenReview] [Video] 

Beyond Research