Relative Probability

Relative probability research focuses on understanding how probabilities are represented and computed, particularly in complex systems like the brain and large language models (LLMs). Current research investigates various probabilistic computation models (e.g., probabilistic population codes, distributed distributional codes, neural sampling codes) and explores how these models relate to reasoning and decision-making, often using Bayesian methods and statistical model checking. This work is significant for advancing our understanding of cognition, improving AI systems' reasoning capabilities, and developing more robust and reliable methods for decision-making under uncertainty in various fields.