Multi-Agent Learning - Lancelot Da Costa
Analysis
This article introduces Lancelot Da Costa, a PhD candidate researching intelligent systems, particularly focusing on the free energy principle and active inference. It highlights his academic background and his work on providing mathematical foundations for the principle. The article contrasts this approach with other AI methods like deep reinforcement learning, emphasizing the potential advantages of active inference for explainability. The article is essentially a summary of a podcast interview or discussion.
Key Takeaways
- •Lancelot Da Costa is a PhD candidate researching intelligent systems, focusing on the free energy principle.
- •He aims to provide mathematical foundations and proofs for the free energy principle.
- •The article contrasts the free energy/active inference approach with other AI methods like deep reinforcement learning.
- •Active inference is presented as potentially advantageous for explainability.
“Lance Da Costa aims to advance our understanding of intelligent systems by modelling cognitive systems and improving artificial systems. He started working with Karl Friston on the free energy principle, which claims all intelligent agents minimize free energy for perception, action, and decision-making.”