Shujian Yu (余书剑)

I am Shujian Yu, an assistant professor of machine learning in the Department of Artificial Intelligence at the Vrije Universiteit Amsterdam. I also hold a guest position in the Machine Learning Group at the UiT - The Arctic University of Norway. I graduated from the University of Florida in the Department of Electrical and Computer Engineering, working with Prof. Jose C. Principe, with a Ph.D. minor in theoretical statistics. Previously, I obtained my BEng degree from the School of Electronic Information and Communications at the Huazhong University of Science and Technology. I was a research scientist of machine learning at the NEC Laboratories Europe from 2019 to 2021.

I am doing research on the intersections between machine learning, information theory and signal processing, which include topics like information-theoretic learning and machine learning for signal processing. Particularly, I am interested in information-theoretic quantities (such as entropy, mutual information, divergence, etc.) estimation, improving the explainability and generalization of deep neural networks by information-theoretic principles (such as the information bottleneck and the principle of relevant information). I am also interested in brain data analysis (such as EEG and fMRI).

I am the recipient of the 2020 International Neural Networks Society Aharon Katzir Young Investigator Award. I am also selected for the 2023 AAAI New Faculty Highlights.

Email  /  Google Scholar  /  GitHub 

profile photo
News

  • 02-05-2024: One paper on neural network-based Granger causality is accepted by International Conference on Machine Learning (ICML).
  • 26-04-2024: One paper on Cauchy-Schwarz divergence for domain adaptation is accepted by International Conference on Uncertainty in Artificial Intelligence (UAI).
  • 27-03-2024: One paper on prototype learning for interpretable brain network-based psychiatric diagnosis and subtyping is accepted by Neuroimage, link, code.
  • 23-01-2024: One paper on Granger causality for interpretable brain network-based psychiatric diagnosis is accepted by Neural Networks, link, code.
  • 16-01-2024: Two papers on “Information Bottleneck”, “Cauchy-Schwarz divergence”, and “generalization error bound” are accepted by International Conference on Learning Representations (ICLR), link1, link2.
  • 15-09-2023: One paper on “Information Bottleneck meets Invariant Risk Minimization” for fine-grained image classification is accepted by Computer Vision and Image Understanding, link.
  • 15-07-2023: One paper on “minimum error entropy” for transfer learning is accepted by ECAI-23, link.
  • 13-06-2023: The slides of our "Information Theory meets Deep Learning" tutorial in ICASSP-23 are available at link.
  • 17-03-2023: One paper accepted by the IEEE Transactions on Information Theory.
  • 27-01-2023: One paper accepted by the IEEE Transactions on Knowledge and Data Engineering.
  • 27-12-2022: One paper accepted by the IEEE Transactions on Signal Processing.
  • 15-12-2022: We will deliver a tutorial entitled “Information Theory meets Deep Learning” in ICASSP-23. See you in Rhodes Island, Greece!
  • 22-11-2022: I am honored to be selected for the 2023 AAAI New Faculty Highlights.
  • 19-11-2022: Two papers accepted by AAAI-23.

Research
method

**Matrix-based Entropy Functional**
Multivariate Extension of Matrix-based Renyi's α-order Entropy Functional
Shujian Yu, Luis Gonzalo Sanchez Giraldo, Robert Jenssen, Jose C. Principe
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020 | code
Measuring Dependence with Matrix-based Entropy Functional
Shujian Yu, Francesco Alesiani, Xi Yu, Robert Jenssen, Jose C. Principe
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2021 | code
Measuring the Discrepancy between Conditional Distributions: Methods, Properties and Applications
Shujian Yu, Ammar Shaker, Francesco Alesiani, Jose C. Principe
Proceedings of the International Joint Conferences on Artificial Intelligence (IJCAI), 2020 | code
Computationally Efficient Approximations for Matrix-Based Rényi's Entropy
Tieliang Gong, Yuxin Dong, Shujian Yu, Bo Dong
IEEE Transactions on Signal Processing, 2022
Robust and Fast Measure of Information via Low-rank Representation
Yuxin Dong, Tieliang Gong, Shujian Yu, Hong Chen, Chen Li
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2023 | code

**Cauchy-Schwarz Divergence**
The Conditional Cauchy-Schwarz Divergence with Applications to Time-Series Data and Sequential Decision Making
Shujian Yu, Hongming Li, Sigurd Løkse, Robert Jenssen, Jose C. Principe
Cauchy-Schwarz Divergence Information Bottleneck for Regression
Shujian Yu, Xi Yu, Sigurd Løkse, Robert Jenssen, Jose C. Principe
Proceedings of the International Conference on Learning Representations (ICLR), 2024 | code
Domain Adaptation with Cauchy-Schwarz Divergence
Wenzhe Yin,Shujian Yu, Yicong Lin, Jie Liu, Jan-Jakob Sonke, Stratis Gavves
Proceedings of the Uncertainty in Artificial Intelligence (UAI), 2024

**Principle of Relevant Information**
Principle of Relevant Information for Graph Sparsification
Shujian Yu, Francesco Alesiani, Wenzhe Yin, Robert Jenssen, Jose C. Principe
Proceedings of the Uncertainty in Artificial Intelligence (UAI), 2022 | code
Multiscale Principle of Relevant Information for Hyperspectral Image Classification
Yantao Wei, Shujian Yu, Luis Gonzalo Sanchez Giraldo, Jose C. Principe
Machine Learning, 2021 | code

**Deep Neural Networks Explainability (Opening the Black-Box of Deep Learning)**
Understanding Autoencoders with Information Theoretic Concepts
Shujian Yu, Jose C. Principe
Neural Networks, 2019 | code
On Kernel Method–based Connectionist Models and Supervised Deep Learning without Backpropagation
Shiyu Duan, Shujian Yu, Yunmei Chen, Jose C. Principe
MIT Neural Computation, 2020 | code
Understanding Convolutional Neural Networks with Information Theory: An Initial Exploration
Shujian Yu, Kristoffer Wickstrøm, Robert Jenssen, Jose C. Principe
IEEE Transactions on Neural Networks and Learning Systems, 2020 | code1 (MATLAB) | code2 (Python)

**Deep Neural Networks Generalization**
Measuring Dependence with Matrix-based Entropy Functional
Shujian Yu, Francesco Alesiani, Xi Yu, Robert Jenssen, Jose C. Principe
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2021 | code
Deep Deterministic Information Bottleneck with Matrix-based Entropy Functional
Xi Yu, Shujian Yu, Jose C. Principe
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2021 | code

**Brain Data Analysis**
Causal Recurrent Variational Autoencoder for Medical Time Series Generation
Hongming Li, Shujian Yu, Jose C. Principe
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2023 | code
BrainIB: Interpretable Brain Network-based Psychiatric Diagnosis with Graph Information Bottleneck
Kaizhong Zheng, Shujian Yu, Baojuan Li, Robert Jenssen, Badong Chen
CI-GNN: A Granger Causality-Inspired Graph Neural Network for Interpretable Brain Network-Based Psychiatric Diagnosis
Kaizhong Zheng, Shujian Yu, Badong Chen

Great template from Jon Barron