Projects
World models
World models for evaluating and interpreting AI agents: their limits, their modular structure, their abstractions, and the channel they model. Continue reading World models
The internal model principle and good regulators
When a controller or agent can be said to model its environment, from the internal model principle to good regulators and possibilistic belief updating. Continue reading The internal model principle and good regulators
Causality
Connecting causal cognition in animals with causal machine and reinforcement learning through disentangled representations. Continue reading Causality
Boundaries
Boundaries between agents and environments, the interfaces they define, and how an agent’s identity is preserved when they change. Continue reading Boundaries
Combining Physical and Intentional Stance for Safe AI
ARIA-funded work connecting physical dynamics with intentional descriptions of beliefs and goals to support verifiable, safer AI. Continue reading Combining Physical and Intentional Stance for Safe AI
Hybrid life
How can we best understand life and cognition? How can we combine the results of different technological advances with natural organisms? Continue reading Hybrid life
Agents and agency
How do we go from “simple physics” to agency? Continue reading Agents and agency
What is uncertainty and how do agents deal with it?
How do we cope with uncertainty? How do we take a decision when we can’t know precisely what’s happening around us, for example, on a misty day? Continue reading What is uncertainty and how do agents deal with it?
