Diptesh Das

Bio

I am a researcher in the Department of Computational Biology and Medical Sciences at the University of Tokyo. My research focuses on statistical machine learning and its application in high-stakes decision-making problems. After finishing my PhD at the University of Tokyo, I worked as a post-doctoral researcher at the University of Tokyo, as a project assistant professor at the Nagoya Institute of Technology, and as a researcher at Nagoya University in Japan. My PhD thesis concentrated on interpretable machine learning models for medical data. In addition to my academic expertise, I have extensive industry experience. After earning a BSc in Physics and a BTech in Applied Physics from the University of Calcutta, India, I began my professional career at TATA Consultancy Services Ltd. in Kolkata, where I worked for several years. I later moved to the UK to pursue an MSc in Advanced Computing from the University of Bristol, where I embarked on a new journey in academia.

Academic Advisors

I have been fortunate to work with the following researchers at different stages of my professional career: Professor Koji Tsuda (PhD Advisor), Professor Ichiro Takeuchi (Postdoc Advisor), Professor Julian Gough (MSc Advisor), Professor Amit Konar (Research Advisor at TATA Consultancy Services Ltd.).

Education

• PhD, Computational Biology and Medical Sciences, The University of Tokyo, Japan.
• MSc, Advanced Computing (Distinction), University of Bristol, Uk.
• BTech, Applied Physics (First Class), University of Calcutta, India.
• BSc, Physics (First Class with Honors), University of Calcutta, India.

Work Experience

• Postdoc Researcher, The University of Tokyo.
• Postdoc Researcher, Nagoya University.
• Project Assistant Professor, Nagoya Institute of Technology.
• Postdoc Researcher, The University of Tokyo.
• IT Analyst, TATA Consultancy Services Ltd., India.

Research Funding

My research is partially funded by the JSPS KAKENHI Young Scientist Research Grant for FY2023-2025.

Corporate Award

I was awarded twice in two consecutive years at TATA Consultancy Services Ltd. One for the outstanding paper award by the CTO of TATA Consultancy Services Ltd. and the other for significant contribution towards intellectual property creation by the Corporate IPR group of TATA Consultancy Services Ltd.

Teaching and Mentorship

I have been actively involved in mentoring and guiding the research projects of both master and graduate students. During my tenure at the Nagoya Institute of Technology, I provided an online lecture for graduate students on the topic of Trustworthy AI for High-Stakes Decision-Making Problems.

Patents

• Method and system for implementation of an interactive television application
  D. Das, A. Ghose, P. Sinha, P. Biswas
  US Patent 9,185,462
• Method and system for automatic tagging in television using crowd sourcing technique
  P. Sinha, R.K. Gupta, A. Ghose, B. Chirabarata, D. Das
  US Patent 14/125,011
• Identification of people using multiple skeleton recording devices
  K. Chakravarty, A. Sinha, D. Das, R. Banerjee, A. Konar, S. Dutta
  US Patent 9,208,376
• System and method for evaluating a cognitive load on a user corresponding to a stimulus
  D. Das, D. Chatterjee, A. Sinharay, A. Sinha
  US Patent 10,827,981
• Selection of electroencephalography (EEG) channels valid for determining cognitive load of a subject
  A. Sinharay, A. Sinha, D. Das, D. Chatterjee
  US Patent 10,499,823

Selected Publications

1. Statistically Robust Sparse High-order Interaction Model
   Diptesh Das*, Ichiro Takeuchi, Koji Tsuda
   AAAI-26 Main Technical Track, 2026.
   [PDF]
   * Work-in-progress version accepted at UAI 2025 Workshop (TPM).
2. DT-sampler: A SAT-based Decision Tree Ensemble
   Xiaotian Xue, Chao Huang, Koji Tsuda, Diptesh Das*
   ACM SIGKDD Explorations, 2025.
   [PDF]
   * Work-in-progress version accepted at UAI 2025 Workshop (Safe AI) and also at ICML 2023 Workshop (IMLH).
3. CRYSIM: Prediction of Symmetric Structures of Large Crystals with GPU-based Ising Machines
   Chen Liang, Diptesh Das, Jiang Guo, Ryo Tamura, Zetian Mao, Koji Tsuda
   Communications Physics, Nature, 2025.
   [PDF]
4. Molecule Graph Networks with Many-body Equivariant Interactions
   Zetian Mao, Chuan-Shen Hu, Jiawen Li, Chen Liang, Diptesh Das, Masato Sumita, Kelin Xia, Koji Tsuda
   Journal of Chemical Theory and Computation, 2025.
   [PDF]
5. A confidence machine for sparse high-order interaction model
   Diptesh Das*, Eugene Ndiaye, Ichiro Takeuchi
   Stat, 2024.
   [PDF]
6. Fast and More Powerful Selective Inference for Sparse High-order Interaction Model
   Diptesh Das*, Vo Nguyen Le Duy, Hiroyuki Hanada, Koji Tsuda, Ichiro Takeuchi
   AAAI 2022 Main Technical Track.
   [PDF]
7. An interpretable machine learning model for diagnosis of Alzheimer's disease
   Diptesh Das*, Junichi Ito, Tadashi Kadowaki, Koji Tsuda
   PeerJ 7 (2019), e6543.
   [PDF]
8. Feature selection by Differential Evolution algorithm -- A case study in personnel identification
   Kingshuk Chakravarty, Diptesh Das, Aniruddha Sinha, Amit Konar
   IEEE Congress on Evolutionary Computation, 2013.
   [PDF]
9. Stabilization of cluster centers over fuzziness control parameter in component-wise Fuzzy c-Means clustering
   Diptesh Das*, Aniruddha Sinha, Kingshuk Chakravarty, Amit Konar
   IEEE International Conference on Fuzzy Systems (FUZZ), 2013.
   [PDF]

Preprints

1. Inverse Design of Metamaterials with Manufacturing-Guiding Spectrum-to-Structure Conditional Diffusion Model
   Jiawen Li, Jiang Guo, Yuanzhe Li, Zetian Mao, Jiaxing Shen, Tashi Xu, Diptesh Das, Jinming He, Run Hu, Yaerim Lee, Koji Tsuda, Junichiro Shiomi
   arXiv:2506.07083 (2025).
   [PDF]
2. Preference-Optimized Pareto Set Learning for Blackbox Optimization
   Haishan Zhang, Diptesh Das*, Koji Tsuda
   arXiv:2408.09976 (2024).
   [PDF]

Peer Review Experience

I review papers for leading machine learning conferences. In 2022, the ICML organizing committee recognized me as one of the top 10% of reviewers of ICML'2022 for my significant contributions as a meta-reviewer.