Jose Marie Antonio Minoza

Researcher in mathematical foundations of AI and Scientific ML. Using dynamical systems theory (Neural Tangent Kernels, Koopman operators, tropical geometry), I develop interpretable, robust physics-informed models for climate science, bioinformatics, and chemistry. Focus: mechanistic interpretability, uncertainty quantification, and responsible AI.

Research Interests

Scientific Machine Learning: Developing interpretable and reliable models for scientific discovery, including physics-informed neural networks, knowledge-guided ML, and hybrid physics-ML approaches. Focus on AI for Science (AI4Sci) and Mathematics for AI (Math4AI).

Mathematical & Computational Modeling: Traditional approaches including differential equations (ODEs, PDEs), agent-based modeling, cellular automata, and numerical methods for simulating complex systems.

Uncertainty Quantification: Probabilistic models, Bayesian methods, Gaussian processes, conformal prediction, and stochastic dynamical systems for quantifying uncertainty in scientific computing.

Mechanistic Interpretability: Understanding neural networks through dynamical systems theory, Neural Tangent Kernels, and learning theory.

Earth & Climate Sciences: Hybrid physics-ML models for environmental hazards, climate modeling, and geoscience applications.

Computational Biology: Transfer learning for biomarker discovery, medical imaging analysis, and healthcare prediction systems.

AI Safety & Alignment: Research on imitation learning, AI alignment, and interpretability for safer AI systems.

Recent News

  • June 2026: Attending Uncertainty Quantification for Climate Summer School (EPSRC MFC CDT and iMPT)
  • May 2026: Paper accepted at ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2026) - Scalable UQ for Extreme Weather Forecasting
  • May 2026: Awarded Gold Reviewer at International Conference on Machine Learning (ICML 2026)
  • May 2026: Paper accepted at International Conference on Machine Learning (ICML 2026) - UltraLIF
  • March 2026: Oral Presentation at Asia-Pacific Summer School and Conference on Networks and Complex Systems, Nanyang Technological University, Singapore
  • January 2026: Oral Presentation at AAAI 2026 Workshop (AI4ES) - ML-EcoLyzer
  • January 2026: Released ML-EcoLyzer Python package on PyPI
  • January 2026: Paper accepted at The Third Conference on Parsimony and Learning (CPAL) - SPIKE
  • December 2025: Two papers accepted at AAAI 2026 Workshops (AI4ES and RSD)
  • November 2025: Speaker on Uncertainty Quantification at R Users Group Philippines
  • November 2025: Speaker at Training on AI in R&D for DOST Researchers
  • September 2025: Selected as ELLIS Summer School Scholar for AI for Earth and Climate Sciences
  • August 2025: Accepted to IAIFI PhD Summer School at MIT
  • August 2025: Accepted to Oxford Machine Learning Summer School (MLx)
  • June 2025: Awarded Environmental AI Research Excellence by AI Safety ATLAS, French Center for AI Safety
  • January 2025: Started School of Climate Change (Hilary Term) at Oxford Climate Society, University of Oxford
  • March 2025: Paper accepted at ICLR 2025 Workshop on Tackling Climate Change with Machine Learning

Selected Awards

  • Gold Reviewer (2026) - International Conference on Machine Learning (ICML 2026)
  • Environmental AI Research Excellence (2025) - AI Safety ATLAS, French Center for AI Safety (CeSIA)
  • ELLIS Summer School Scholar (2025) - ELLIS Unit Jena - AI for Earth and Climate Sciences
  • Magna Cum Laude (2016) - BSc Applied Mathematics with IT, Far Eastern University Manila (First Class Honours equivalent)
  • LEAP Scholarship (2014) - Long-Term Educational Assistance Program, Far Eastern University Manila
  • DOST National Invention Contest and Exhibits (2012) - Finalist, Featured in GMA 7 iBilib
  • DOST Regional Invention Contest and Exhibits - Creative Research Category (SIBOL Award) (2011) - Region IV-A Winner, National Finalist presented at NSTW
  • Outstanding Research Award (2011) - Rizal National Science High School