Tianjian Huang
 |
PhD Candidate |
About me
I am currently a Ph.D. candidate at University of Southern California and I am fortunate enough to be advised by Prof. Meisam Razaviyayn. Prior to that, I received my B.Sc. degree from Rensselaer Polytechnic Institute.
Research
My research interests lie in:
Scalable training and fine-tuning methodologies for robust and responsible AI
Publications
[ICML 2024] Optimal Differentially Private Model Training with Public Data. [code]
Andrew Lowy, Zeman Li, Tianjian Huang, Meisam Razaviyayn.
[ICML 2022]A Rigorous Study of Integrated Gradients Method and Extensions to Internal Neuron Attributions. [code]
Daniel D Lundstrom, Tianjian Huang, Meisam Razaviyayn.
[TMLR 2022] Robustness Through Data Augmentation Loss Consistency. [code]
Tianjian Huang, Shaunak Halbe, Chinnadhurai Sankar, Pooyan Amini, Satwik Kottur, Alborz Geramifard, Meisam Razaviyayn, Ahmad Beirami.
[AISTATS 2021] Alternating Direction Method of Multipliers for Quantization. [code]
Tianjian Huang, Prajwal Singhania, Maziar Sanjabi, Pabitra Mitra, Meisam Razaviyayn.
[Signal Processing 2021] A Decentralized Adaptive Momentum Method for Solving a Class of Min-Max Optimization Problems.
Babak Barazandeh, Tianjian Huang, George Michailidis.
[NeurIPS 2019] Solving a Class of Non-convex Min-Max Games Using Iterative First Order Methods. [code]
Nouiehed, Maher, Maziar Sanjabi, Tianjian Huang, Jason D. Lee, and Meisam Razaviyayn.
[IEEE Signal Processing Magazine 2019] Non-convex Min-Max Optimization: Applications, Challenges, and Recent Theoretical Advances.
Razaviyayn, Meisam, Tianjian Huang, Songtao Lu, Maher Nouiehed, Maziar Sanjabi, and Mingyi Hong.