He Zhao

Greetings! This is my research page.

I’m a machine learning researcher at CSIRO’s Data61 and an adjunct research fellow at Monash University.

I’m interested in probabilistic approaches that improve robustness, generalisation, uncertainty estimation, and interpretation of machine/deep learning. I’m also into applications of optimal transport and Bayesian statistics (e.g., probabilistic models and variational inference) in the deep learning context.

🏆 Our paper led by Vy Vo won Best Student Paper - Research at KDD 2023.

Selected publications

[ICML 2024] Optimal Transport for Structure Learning Under Missing Data
V. Vo, H. Zhao, T. Le, E. V. Bonilla, D. Phung
link, code
[ICML 2024] Parameter Estimation in DAGs from Incomplete Data via Optimal Transport
V. Vo, T. Le, L. T. Vuong, H. Zhao, E. V. Bonilla, D. Phung
link, code
[ICML 2024] Distribution Alignment Optimization through Neural Collapse for Long-tailed Classification
J. Gao, H. Zhao, D. Guo, H Zha
link, code
[ICLR 2024] PTaRL: Prototype-based Tabular Representation Learning via Space Calibration
H. Ye, W. Fan, X. Song, S. Zheng, H. Zhao, D. Guo, Y. Chang
Spotlight, top 5%
link
[NeurIPS 2023] Beyond Unimodal: Generalising Neural Processes for Multimodal Uncertainty Estimation
M.C. Jung, H. Zhao, J. Dipnall, L. Du
link
[NeurIPS 2023] NPCL: Neural Processes for Uncertainty-Aware Continual Learning
S. Jha, D. Gong, H. Zhao, L. Yao
link, code
[NeurIPS 2023] Enhancing Minority Classes by Mixing: An Adaptative Optimal Transport Approach for Long-tailed Classification
J. Gao, H. Zhao, Z. Li, D. Guo
link, code
[KDD 2023] Feature-based Learning for Diverse and Privacy-Preserving Counterfactual Explanations
V. Vo, T. Le, V. Nguyen, H. Zhao, E. Bonilla, G. Haffari, D. Phung
🏆 Best Student Paper - Research, link, code
[TMLR 2023] Generating Adversarial Examples with Task Oriented Multi-Objective Optimization
A. Bui, T. Le, H. Zhao, Q.H. Tran, P. Montague, D. Phung
Transactions on Machine Learning Research, link, code
[ICML 2023] Transformed Distribution Matching for Missing Value Imputation
H. Zhao, K. Sun, A. Dezfouli, E. Bonilla
link, code
[ICML 2023] Vector Quantized Wasserstein Auto-Encoder
L.T. Vuong, T. Le, H. Zhao, C. Zheng, M. Harandi, J. Cai, D. Phung
link, code
[NeurIPS 2022] Uncertainty Estimation for Multi-view Data: The Power of Seeing the Whole Picture
M.C. Jung, H. Zhao, J. Dipnall, B. Gabbe, L. Du
link
[NeurIPS 2022] Adaptive Distribution Calibration for Few-Shot Learning with Hierarchical Optimal Transport
D. Guo, L. Tian, H. Zhao, M. Zhou, H. Zha
link, code
[NeurIPS 2022] Learning to Re-weight Examples with Optimal Transport for Imbalanced Classification
D. Guo, Z. Li, M. Zheng, H. Zhao, M. Zhou, H. Zha
link, code
[ICLR 2022] A Unified Wasserstein Distributional Robustness Framework for Adversarial Training
A. Bui, T. Le, Q. Tran, H. Zhao, D. Phung
link, code
[ICLR 2022] Representing Mixtures of Word Embeddings with Mixtures of Topic Embeddings
D. Wang, D. Guo, H. Zhao, H. Zheng, K. Tanwisuth, B. Chen, M. Zhou
link, code
[AISTATS 2022] A Global Defense Approach via Adversarial Attack and Defense Risk Guaranteed Bounds
T. Le, A. Bui, M.T.T. Le, H. Zhao, P. Montague, Q. Tran, D. Phung
link
[AISTATS 2022] Particle-based Adversarial Local Distribution Regularization
T. Nguyen, T. Le, H. Zhao, J. Cai, D. Phung
link
[IJCAI 2021 Survey Track] Topic Modelling Meets Deep Neural Networks: A Survey
H. Zhao, D. Phung, V. Huynh, Y. Jin, L. Du, W. Buntine
link, arxiv
[ICLR 2021] Neural Topic Model via Optimal Transport
H. Zhao, D. Phung, V. Huynh, T. Le, W. Buntine
Spotlight, top 5%
link, code
[NeurIPS 2020] OTLDA: A Geometry-aware Optimal Transport Approach for Topic Modeling
V. Huynh, H. Zhao, D. Phung
link
[ECCV 2020] Improving Adversarial Robustness by Enforcing Local and Global Compactness
A. Bui, T. Le, H. Zhao, P. Montague, O. de Vel, T. Abraham, D. Phung
link, arxiv, code
[AISTATS 2020] Variational Autoencoders for Sparse and Overdispersed Discrete Data
H. Zhao, P. Rai, L. Du, W. Buntine, D. Phung, M. Zhou
link, code
[NeurIPS 2018] Dirichlet Belief Networks for Topic Structure Learning
H. Zhao, L. Du, W. Buntine, M. Zhou
link, code
[ICML 2018] Inter and Intra Topic Structure Learning with Word Embeddings
H. Zhao, L. Du, W. Buntine, M. Zhou
link, code
[AISTATS 2018] Bayesian multi-label learning with sparse features and labels, and label co-occurrences
H. Zhao, P. Rai, L. Du, W. Buntine
link, code
[ACML 2017] A Word Embeddings Informed Focused Topic Model
H. Zhao, L. Du, W. Buntine
link, code
[ICDM 2017] MetaLDA: A Topic Model that Efficiently Incorporates Meta information
H. Zhao, L. Du, W. Buntine, G. Liu
link, arxiv, code
[ICML 2017] Leveraging Node Attributes for Incomplete Relational Data
H. Zhao, L. Du, W. Buntine
link, arxiv, code

Professional Services

Reviewing
NeurIPS (A top 30% reviewer in 2018), ICML, ICLR, JMLR, JAIR, TPAMI, IJCV, TKDE, Machine Learning Journal (Editorial board member), ACML (Area Chair) …
Tutorials
Optimal Transport for Unsupervised Learning at Asian Conference on Machine Learning 2021

Recent Talks

Enhancing Generalisation and Robustness in Deep Learning with Optimal Transport
School of Computer Science and Engineering, UNSW Sydney, July 2024
slides
Learning Deep Representations with Optimal Transport
Statistics Seminar, The University of Sydney
link