Positions

• Head, Department of Data Science and AI, Faculty of Information Technology, Monash University, Australia (2024 - now)
• Research Director, Department of Data Science and AI, Faculty of Information Technology, Monash University, Australia (2020 - 2023)
• Professor of Machine Learning and Data Science, Faculty of Information Technology, Monash University, Australia.
• Senior Principal Scientist/Consultant, VinAI Research/VinFast Australia (2021-now).
• AI Chief Scientist/Consultant, Trusting Social aka Credit AI, Melbourne, Australia (2017-2020).
• I'm also affiliated with the Machine Learning group and the Digital Health Initiatives at Monash.

Prospective PhD students, Postdocs and Tutors: Thank you for your interest and for reaching out to me. While I'm keen on receiving strong applications and your EoI, due to the large volume of such emails, please accept my apology in advance if you do not receive a response from me individually as I might only be able to respond to short-listed or selected EoI. Please see this link for further information regarding PhD scholarship and admission at Monash.

Research Interest

• Theoretical Foundations of Machine Learning and AI
• Optimal Transport and Wasserstein space methods in AI
• Generative AI and LLMs
• Robust/Adversarial Machine Learning and Trustworthy AI
• Revisiting Graphical Models and Causality in the Deep Learning era

Earlier research interests

• Nonparametric Machine Learning: Bayesian Nonparametrics, Random Finite Sets and Point Process for ML
• Graphical Models, Probabilistic Inference, Representation Learning, Latent Variable Models
• Optimisation, Online learning, Learning from Non-stationary Distributions
• Applications: Medical AI and Digital Health, NLP, Cybersecurity and Digital Identity, Computer vision, AI-enabled Autism Research

Some Recent Services

• With Geoff Webb and Claude Sammut, we are working on the third/living edition for the Encyclopedia of Machine Learning and Data Science . Please drop us an email if you would like your work to be considered for this edition. Here is the link to the second edition for your reference.
• From 2021, I am an Associate Editor for the Journal of Artificial Intelligence Research (JAIR)
• Senior PC: 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2021
• Area Chair: IEEE Int. Conference on Data Mining (ICDM), 2021
• TPC Member: Thirty-third Conference on Neural Information Processing Systems (NeurNIPS), 2020
• Area Chair: Asian Conference on Machine Learning (ACML), 2020
• TPC Member: International Conference on Machine Learning (ICML), 2019
• TPC Member: Thirty-third Conference on Neural Information Processing Systems (NeurNIPS), 2019
• TPC Member: AAAI Conference on Artificial Intelligence (AAAI), 2019
• Senior PC Member: Asian Conference on Machine Learning (ACML), 2019
• Program Chair: The 22nd Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD18), June 3 - 6, 2018, Melbourne Australia.
• TPC Member: International Conference on Machine Learning (ICML), 2018
• TPC Member: ACM SIGKDD Conference on Knowledge Discovery and Data Mining (SIGKDD), 2018
• Senior PC Member: International Joint Conference on Artificial Intelligence (IJCAI), 2018
• Reviewer: The 6th International Conference on Learning Representations (ICLR), 2018
• TPC Member: IEEE International Conference on Data Mining (ICDM), 2018
• TPC Member: Annual Conference on Neural Information Processing Systems (NIPS), 2017
• TPC Member: International Conference on Machine Learning (ICML), 2017
• TPC Member: ACM SIGKDD Conference on Knowledge Discovery and Data Mining (SIGKDD), 2017
• Senior PC Member: Asian Conference on Machine Learning (ACML), 2017
• TPC Member: Asia-Pacific Conference on Knowledge Discovery and Data Mining (PAKDD), 2017
• Senior PC Member: International Joint Conference on Artificial Intelligence (IJCAI), 2016
• TPC Member: ACM SIGKDD Conference on Knowledge Discovery and Data Mining (SIGKDD), 2016
• Senior PC Member: Asian Conference on Machine Learning (ACML), 2016
• TPC Member: Asia-Pacific Conference on Knowledge Discovery and Data Mining (PAKDD), 2016
• TPC Member: IEEE Int. Conf. on Pervasive Computing and Communications (PERCOM), 2016
• TPC Member: Annual Conference on Neural Information Processing Systems (NIPS), 2015
• TPC Member: ACM SIGKDD Conference on Knowledge Discovery and Data Mining (SIGKDD), 2015
• TPC Member: International Joint Conference on Artificial Intelligence (IJCAI), 2015
• Tutorial Co-Chair: Asia-Pacific Conference on Knowledge Discovery and Data Mining (PAKDD), 2015
• TPC Member: Asia-Pacific Conference on Knowledge Discovery and Data Mining (PAKDD), 2015
• Senior PC Member: Asian Conference on Machine Learning (ACML), 2015
• TPC Member: IEEE Int. Conf. on Pervasive Computing and Communications (PERCOM), 2015
• Program Chair (with Hang Li), Asian Conference on Machine Learning (ACML), 2014.
• TPC Member: ACM Conference on Multimedia (ACM-MM), 2014

Recent News

• Bridging Global Context Interactions for High-Fidelity Pluralistic Image Completion accepted to Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
• Parameter Estimation in DAGs from Incomplete Data via Optimal Transport accepted to ICML 2024
• Optimal Transport for Structure Learning Under Missing Data accepted to ICML 2024
• NAYER: Noisy Layer Data Generation for Efficient and Effective Data-free Knowledge Distillation accepted to CVPR 2024
• Taming Stable Diffusion for Text to 360 Panorama Image Generation accepted to CVPR 2024
• Text-Enhanced Data-free Approach for Federated Class-Incremental Learning accepted to CVPR 2024
• Revisiting Deep Audio-Text Retrieval Through the Lens of Transportation accepted to ICLR 2024
• I'm deeply honoured to take the Head of Department of Data Science and AI role from Jan 2024.
• I'm very happy to receive the Dean Award for Excellent in Research, 2023 - this is particularly meaningful for me when it came from my own wonderful colleagues' nomination.
• In case you are interested in the implication of GenAI and its implication, this is my recent talk AI and Biometrics in the Age of Generative Intelligence at Victorian Parliament Library and Australian Academy of Technological Sciences and Engineering (ATSE), 19 October, 2023. (slides)
• Model and Feature Diversity for Bayesian Neural Networks in Mutual Learning accepted to NeurIPS 2023
• Systematic Assessment of Factual Knowledge in Large Language Models accepted to EMNLP 2023 (Findings)
• Flat Seeking Bayesian Neural Networks accepted to NeurIPS 2023
• Optimal Transport Model Distributional Robustness accepted to NeurIPS 2023
• Our PhD student, Vy Vo, won the Best Student Paper Award at SIGKDD 2023 for the paper Feature-based Learning for Diverse and Privacy-Preserving Counterfactual Explanations
• You might be interested in my recent talk at AI@MMelbourne Colloquium on the topic of Trustworthy AI and Optimal Transport (slide is here)
• Vector Quantized Wasserstein Auto-Encoder, accepted to ICML 2023
• An Additive Instance-Wise Approach to Multi-class Model Interpretation, accepted to ICLR 2023
• Global-Local Regularization Via Distributional Robustness, accepted to AISTATS 2023
• Feature-based Learning for Diverse and Privacy-Preserving Counterfactual Explanations, accepted to SIGKDD 2023
• On Cross-Layer Alignment for Model Fusion of Heterogeneous Neural Networks, accepted to ICASSP 2023 (with UT Austin and VinAI)
• Cross-adversarial local distribution regularization for semi-supervised medical image segmentation, accepted to MICCAI 2023
• Stochastic Multiple Target Sampling Gradient Descent, accepted to NeurIPS 2022
• MoVQ: Modulating Quantized Vectors for High-Fidelity Image Generation (spotlight), accepted to NeurIPS 2022 (project page)
• On Transportation of Mini-batches: A Hierarchical Approach, accepted to ICML 2022
• Bridging Global Context Interactions for High-Fidelity Image Completion, accepted to CVPR 2022
• Domain Generalisation of NMT: Fusing Adapters with Leave-One-Domain-Out Training accepted to ACL 2022
• A Unified Wasserstein Distributional Robustness Framework for Adversarial Training accepted to ICLR 2022
• Cycle Class Consistency with Distributional Optimal Transport and Knowledge Distillation for Unsupervised Domain Adaptation, accepted to UAI 2022
• A Global Defense Approach via Adversarial Attack and Defense Risk Guaranteed Bounds accepted to AISTATS 2022
• Particle-based Adversarial Local Distribution Regularization accepted to AISTATS 2022
• Sobolev Transport: A Scalable Metric for Probability Measures with Graph Metrics (collaborated with RIKEN, Akron University and VinAI) has been accepted to AISTATS 2022
• If you are curious and/or interested in optimal transport theory and how it can be used in machine learning, you might want to check out a recent tutorial from my research group with Nhat Ho from UT Austin here
• On Efficient Multilevel Clustering via Wasserstein Distances, published in JMLR, 2021
• On Learning Domain-Invariant Representations for Transfer Learning with Multiple Sources (collaborated with VinAI), published in NeurIPS 2021
• Exploiting Domain-Specific Features to Enhance Domain Generalization (collaborated with VinAI), published in NeurIPS 2021
• Improving Kernel Online Learning with a Snapshot Memory has been published in Machine Learning Journal (MLJ)
• Generalised Unsupervised Domain Adaptation of Neural Machine Translation with Cross-Lingual Data Selection, published in EMNLP 2021
• STEM: An Approach to Multi-Source Domain Adaptation with Guarantees, published in ICCV 2021
• LAMDA: Label Matching Deep Domain Adaptation, published in ICML 2021
• TIDOT: A Teacher Imitation Learning Approach for Domain Adaptation with Optimal Transport, published in IJCAI 2021 (research track)
• Topic Modelling Meets Deep Neural Networks: A Survey, published in IJCAI 2021 (survey track)
• MOST: Multi-Source Domain Adaptation via Optimal Transport for Student-Teacher Learning, published in UAI 2021
• Optimal transport for deep generative models: State of the art and research challenges, published in IJCAI 2021 (survey track)
• Neural Topic Model via Optimal Transport, published in ICLR 2021. Check out the code here which also includes Sinkhorn implementation in Tensorflow and Pytorch within a subroutine to learn the topics.
• Improving Ensemble Robustness by Collaboratively Promoting and Demoting Adversarial Robustness, published in AAAI 2021
• I was the Finalist for the 2020 Australian Museum Eureka Prize for Excellence in Data Science (usually dubbed as the 'Oscars' for Australian science)
• OTLDA: A Geometry-aware Optimal Transport Approach for Topic Modeling, published in NeurIPS 2020
• Parameterized Rate-Distortion Stochastic Encoder, published in ICML 2020
• Variational Autoencoders for Sparse and Overdispersed Discrete Data, published in AISTATS 2020
• On Efficient Multilevel Clustering via Wasserstein Distances, published in JMLR (with minor revision)
• Improving Adversarial Robustness by EnforcingLocal and Global Compactness, published in ECCV 2020
• Robust Variational Learning for Multiclass Kernel Models with Stein Refinement, published in TKDE journal
• Effective Unsupervised Domain Adaptation with Adversarially Trained Language Models, published in EMNLP 2020
• Congratulations to PhD student, Quan Hoang, on winning the ACEMS Business Analytics Prize for 2020 from the Australian Centre of Excellence in Mathematical and Statistical Frontiers on his work of using rate-distortion theory for robust machine learning.
• Congratulations to Dr Trung Le, who has recently taken up a new position as Lecturer (Assistant Professor) in the Department in Data Science and AI, Monash University, Australia.
• A Relational Memory-based Embedding Model for Triple Classification and Search Personalization, published in ACL 2020
• Three-Player Wasserstein GAN via Amortised Duality, published in IJCAI 2019
• Learning How to Active Learn by Dreaming, published in ACL 2019
• Learning Generative Adversarial Networks from Multiple Data Sources, published in IJCAI 2019
• [[https://aclanthology.org/N19-1226.pdf | A Capsule Network-based Embedding Model for Knowledge Graph Completion and Search Personalization], published in NAACL 2019
• Robust Anomaly Detection in Videos using Multilevel Representations, published in AAAI 2019
• [[https://proceedings.mlr.press/v89/ho19a.html | Probabilistic Multilevel Clustering via Composite Transportation Distance], published in AISTATS 2019
• Maximal Divergence Sequential Autoencoder for Binary Software Vulnerability Detection, published in ICLR 2019
• Robust Bayesian Kernel Machine via Stein Variational Gradient Descent for Big Data, published in KDD 2018
• MGAN: Training Generative Adversarial Nets with Multiple Generators, published in ICLR 2018
• Learning Graph Representation via Frequent Subgraphs, published in SDM 2018
• GEN: Geometric Enclosing Networks, published in IJCAI 2018
• Dual Discriminator Generative Adversarial Nets, published in NeurIPS 2017
• Supervised Restricted Boltzmann Machines, published in UAI 2017
• Approximation Vector Machines for Large-Scale Online Learning, published in Journal of Machine Learning Research (JMLR 2017)
• Multilevel Clustering via Wasserstein Means, published in ICML 2017
• Our PhD student, Hung Vu, won the Best Application Paper Award (Energy-Based Localized Anomaly Detection in Video Surveillance, PAKDD 2017)
• Discriminative Bayesian Nonparametric Clustering, published in IJCAI 2017
• Large-scale Online Kernel Learning with Random Feature Reparameterization, published in IJCAI 2017
• GoGP: Fast Online Regression with Gaussian Processes, published in ICDM 2017
• Prediction of Population Health Indices from Social Media using Kernel-based Textual and Temporal Features, published in WWW 2017
• Hierarchical semi-Markov conditional random fields for deep recursive sequential data, published in AIJ 2017 (Artificial Intelligence journal).
  • See my thesis (chapter 5) for an equivalent directed graphical model, which is the precusor of this work and where I had described the Assymetric Inside-Outside (AIO) algorithm in great detail. A brief version of this for directed case has also appeared in this AAAI'04's paper. The idea of semi-Markov duration modelling has also been addressed for directed case in these CVPR05 and AIJ09 papers.
• Column Networks for Collective Classification, published in AAAI 2017.
• Dual Space Gradient Descent for Online Learning, published in NIPS 2016.
• Scalable Baysian Nonparametric Multilevel Clustering, published in UAI 2016.
• Budgeted Semi-supervised Support Vecitor Machine, published in UAI 2016.
• Nonparametric Budgeted Stochastic Gradient Descent, published in AISTATS 2016.
• One-pass Logistic Regression for Label-drift and Large-scale Classification on Distributed Systems, published in ICDM 2016.

... more news