dynamic expectation maximisation (DEM)

A variational approach for dynamic Bayesian inversion of nonlinear causal models, named in this paper.

Neighborhood — ranked by edge-count

paper

framework

Free Energy Principle
implements
A foundational variational principle from statistical physics that formalizes how self-organizing systems maintain structural integrity and adapt to their environment by minimizing free energy—a mathematical bound on surprise or prediction error. Originally developed by Karl Friston, the framework unifies action, perception, and learning as processes of active inference, where systems both update internal models of the world and act upon it to reduce the divergence between predictions and observations.
Empirical Bayes
implements
Statistical framework underlying perceptual inference and learning scheme; enables hierarchical models of sensory generation.

method

variational filtering
implements
Method to obtain time-dependent conditional densities by maximizing variational free energy.
hierarchical models
implements
Models of sensory generation that allow dynamic context-sensitive prior expectations.

event

Variational filtering and inference (lecture May 30, 2008)
about
Second lecture presenting variational filtering and DEM as a dynamic Bayesian inversion method.

hypothesis

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

"any goal or purpose can be well thought of as maximization of the expected value of the cumulative sum of a received scalar signal (reward)"quote0.718
The reward hypothesis underpinning RL, quoted from Sutton and Barto.
Dirichlet Parameter Accumulationmethod0.717
Learning rule for updating Dirichlet beliefs about likelihood matrix A by adding outer products of observations and state estimates.
Dynamic Weight Average (DWA)method0.713
Loss balancing based on learning speed.
Multi-Agent Deep Deterministic Policy Gradient (MADDPG)method0.713
RL algorithm used to train baseline agents in the physical deception environment
Coherence Maximizationconcept0.706
The functional role consciousness plays: minimizing constraint violations between simultaneously active partial models of reality
maximum-norm gradient normalizationmethod0.705
Training-free technique normalizing all task gradients to the maximum gradient norm magnitude
Dynamic Exploration Incentiveconcept0.705
The property of the pain-belief signal that adapts to environmental changes rather than providing a fixed exploration bonus
Dynamic dispatchconcept0.702