Resume
Cybernetics | ALife | AI safety researcher. Mathematical modeller. Coder.
Profile
Full CVCross-disciplinary researcher developing foundations of intelligent behaviour for agency in cybernetics, AI, ALife, and cognitive science.
Building mathematical models for AI systems that are interpretable and safe for society.
Coding of complex systems and embodied agents driven by mathematically sound, and safe, definitions and models.
Experience
Lead Researcher | ARIA R&D CreatorOct 2026 – presentAraya Inc.
Previously
Research Team Leader | ARIA R&D CreatorOct 2025 – Sep 2026
Chief ResearcherOct 2022 – Sep 2025
ResearcherNov 2021 – Sep 2022
Board of DirectorsNov 2023 – presentInternational Society for Artificial Life
Senior Visiting Research FellowJan 2023 – presentUniversity of Sussex
Previously
Visiting Research FellowJan 2020 – Dec 2022
Postdoctoral Research FellowMar 2019 – Oct 2019
Lecturer in Machine Learning (Fixed term)Feb 2019 – Jun 2019
Lecturer in Agent Foundations (Fixed term)Dec 2022 – Feb 2023Center for Human Nature, Artificial Intelligence, and Neuroscience (CHAIN), Hokkaido University
Research Consultant and DeveloperOct 2021 – Oct 2021Nested Minds Network Ltd.
JSPS | Royal Society Postdoctoral Research FellowNov 2019 – Oct 2021RIKEN Centre for Brain Science
Education
| Degree | Field | Institution | Dates |
|---|---|---|---|
| Ph.D. | Informatics and Artificial Intelligence | University of Sussex | Oct 2015 – Jul 2019 |
| M.Sc. | Evolutionary and Adaptive Systems | University of Sussex | Oct 2013 – Sep 2014 |
| B.Eng. | Information and Business Organisation Engineering | University of Trento | Oct 2009 – Sep 2012 |
Selected publications
Compositional Behavioral Semantics for State Abstraction in Reinforcement Learning
Zhang, Y., Luo, Z., & Baltieri, M. · International Conference on Machine Learning (ICML)
Publisher · PreprintWorld models of environment, agent and joint agent-environment systems
Baltieri, M., Torresan, F., Zhang, Y., Boyd, A., & Rosas, F. E. · arXiv:2608.20401
PreprintAI in a vat: Fundamental limits of efficient world modelling for agent sandboxing and interpretability
Rosas, F., Boyd, A. & Baltieri, M. · Reinforcement Learning Conference (RLC)
Publisher · PreprintMathematical approaches to the study of agents
Baltieri, M., & Suzuki, K. · Philosophical Transactions of the Royal Society B · Invited submission, to appear
PreprintA "good regulator theorem" for embodied agents
Virgo, N., Biehl, M., Baltieri, M., & Capucci, M. · Artificial Life Conference (ALIFE)
Publisher · PreprintA Bayesian Interpretation of the Internal Model Principle
Baltieri, M., Biehl, M., Capucci, M., & Virgo, N. · arXiv:2503.00511
PreprintDisentangled Representations for Causal Cognition
Torresan, F. & Baltieri, M. · Physics of Life Reviews
Publisher · PDF · PreprintHybrid Life: Integrating Biological, Artificial, and Cognitive Systems
Baltieri, M., Iizuka, H., Witkowski, O., Sinapayen, L., & Suzuki, K. · WIREs Cognitive Science · Invited submission
Publisher · PDF · Preprint