A framework for the use of generative modelling in non-equilibrium statistical mechanics

Paper · arXiv 2406.11630 · Published June 17, 2024
Mechanistic Interpretability

We discuss an approach to mathematically modelling systems made of objects that are coupled together, using generative models of the dependence relationships between states (or trajectories) of the things comprising such systems. This broad class includes open or non-equilibrium systems and is especially relevant to self-organising systems. The ensuing variational free energy principle (FEP) has certain advantages over using random dynamical systems explicitly, notably, by being more tractable and offering a parsimonious explanation of why the joint system evolves in the way that it does, based on the properties of the coupling between system components. Using the FEP allows us to model the dynamics of an object as if it were a process of variational inference, because variational free energy (or surprisal) is a Lyapunov function for its dynamics. In short, we argue that using generative models to represent and track relations among subsystems leads us to a particular statistical theory of interacting systems. Conversely, this theory enables us to construct nested models that respect the known relations among subsystems.

Synthesis notes from this paper's topics 8

This paper has no direct semantic matches yet — here are notes from the topic(s) it's filed under.

Lines of inquiry this paper opens 24

Research framings built by reading the notes related to this paper — the questions it feeds into.

How do neural networks internally organize representations and reasoning? Why does task-specific specialization harm generalization and future learning? Does recurrent computation enable new reasoning capabilities fixed-depth architectures cannot achieve? Do language models genuinely understand and apply human social norms? How do language models organize knowledge across layers and representations? Why do language models produce convincing outputs despite flawed reasoning? How do representation structure properties emerge and change in neural networks? Why do persistent gaps exist between user intent and AI execution? Can human psychological models explain how LLMs actually reason? What determines whether language models adopt prompted personas successfully? How effectively does mechanistic interpretability explain model reasoning and guide design? Do language models learn genuine linguistic structure or surface-level patterns? Does encoded knowledge in language models causally influence their outputs? Do foundation models build genuine world models or task-specific shortcuts? How do different adaptation mechanisms compare on efficiency and capability preservation?