Tetsuya Ishikawa
Everything in the World is Made of Mathematics
About Me
I am a mathematical engineer, research scientist, and team leader with over a decade of experience applying mathematics to engineering challenges. My work spans physical modeling, sensor algorithms, computer vision, deep learning, manufacturing optimization, and Physical AI. I specialize in addressing uncertain technical problems through mathematical formulation, prototyping, validation, and practical application. My experience includes magnetic sensor development, Driver Monitoring Systems, and Physical AI research. I currently lead a research team and contribute to technology strategy. I am particularly interested in first-principles modeling, data-driven methods, and their application to real-world engineering challenges.
Skills
Mathematical Modeling
- Probability and statistics
- Calculus and numerical methods
- Dynamical systems
- First-principles physical modeling
- Sensor modeling and estimation
Machine Learning and Computer Vision
- Deep learning
- Image processing
- Generative models
- Privacy-preserving vision
- Small-data learning
- Human and driver modeling
Engineering and Deployment
- Python, C/C++, MATLAB
- PyTorch, TensorFlow, scikit-learn
- Linux, Git, Docker
- Data analysis and experimental validation
Research and Leadership
- Research planning
- Team leadership and mentoring
- Technology evaluation
- Cross-functional collaboration
- Mid-term technology strategy
Publications
- M. Mabuchi and T. Ishikawa, "Patch-based Privacy Preserving Neural Network for Vision Tasks", Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023.
- T. Ishikawa and S. Stent, "Boosting Supervised Learning in Small Data Regimes with Conditional GAN Augmentation," 2022 IEEE International Conference on Image Processing (ICIP), 2022. (Paper , Live ICIP Poster session )
- T. Ishikawa and T. Hayakawa, "Chaotic Behavior of the Folding Map on the Equilateral Triangle", 9th IFAC Symposium on Nonlinear Control Systems, 2013. (Paper )
- T. Ishikawa and T. Hayakawa, "Analysis of Critical Phenomenon on Gossip Protocol using Back-Ultradiscretization", 9th IFAC Symposium on Nonlinear Control Systems, 2013. (Paper )
- T. Ishikawa and T. Hayakawa, "Extended gossip protocol for diffusion of multiple messages and its percolation probability", 2011 50th IEEE Conference on Decision and Control and European Control Conference, 2011. (Paper )
- T. Ishikawa and T. Hayakawa, "On Equivalence between Critical Probabilities of Dynamic Gossip Protocol and Static Site Percolation", SICE Journal of Control, Measurement, and System Integration, 2011. (Paper )
- T. Ishikawa and T. Hayakawa, "Gossip protocol on the ad hoc networks and its approximated saturation", 49th IEEE Conference on Decision and Control (CDC), 2010. (Paper )
Career
May 2024 – Present
Conduct research on Physical AI, lead a machine learning research team, and contribute to the department’s mid-term management and technology planning.
Jan 2022 – Apr 2024
On secondment from AISIN Corporation, conducted research and early-stage development for Driver Monitoring Systems, focusing on machine learning and image synthesis technologies. The organization was known as Woven CORE, Inc. until March 2023.
Jan 2020 – Dec 2021
On secondment from AISIN Corporation, conducted research and development in image processing and deep learning for Driver Monitoring Systems, and applied data analysis to improve lithium-ion battery manufacturing operations.
Oct 2018 – Dec 2019
On secondment from AISIN Corporation, developed deep-learning-based image processing algorithms for Driver Monitoring Systems.
Feb 2018 – Sep 2018
Conducted research on image processing and deep learning, with a focus on neural network model compression and acceleration.
Apr 2012 – Jan 2018
Developed a corrosion monitoring system based on a magnetic sensor array, focusing on physical models and algorithms for estimating corrosion location, depth, and volume from sensor measurements.
Apr 2006 – Mar 2012
Earned a Bachelor of Engineering in Control and Systems Engineering in March 2010 and a Master of Engineering in Mechanical and Environmental Informatics in March 2012.
Open to Collaboration
I am open to selected consulting, research, and development opportunities.
Feel free to contact me to discuss a potential project or collaboration.