Flat minima are illusory; generalization is driven by weakness, a reparameterization-invariant measure of compatible completions that predicts performance better than sharpness on MNIST and Fashion-MNIST.
J R Soc Interface, 2013; 10 (86): 20130475
6 Pith papers cite this work. Polarity classification is still indexing.
citation-role summary
citation-polarity summary
verdicts
UNVERDICTED 6representative citing papers
Information defined as maximum-caliber deviation derives IIT 3.0 cause-effect repertoires from constrained entropy maximization and equates to prediction error under CLT and LDT.
An active inference POMDP model of sustained reading formalizes three lighting channels and generates six predictions all confirmed in 20 Monte Carlo simulations.
Multi-agent active inference digital twins with contextual inference and streaming ML integration, illustrated via a Cournot competition socio-economic example.
Simulations show that superposed Lotka-Volterra chemical oscillations can bias binary polymer elongation to produce catalytic loops and low-entropy informational sequences at higher rates than random selection.
Active inference model unifies human collision avoidance by reproducing meta-analysis aggregates and simulator-specific effects on response timing, maneuver selection, and execution.
citing papers explorer
-
Are Flat Minima an Illusion?
Flat minima are illusory; generalization is driven by weakness, a reparameterization-invariant measure of compatible completions that predicts performance better than sharpness on MNIST and Fashion-MNIST.
-
Information as Maximum-Caliber Deviation: A bridge between Integrated Information Theory and the Free Energy Principle
Information defined as maximum-caliber deviation derives IIT 3.0 cause-effect repertoires from constrained entropy maximization and equates to prediction error under CLT and LDT.
-
How Light Reshapes the Mind. An Active Inference Framework for the Cognitive and Emotional Effects of Indoor Lighting
An active inference POMDP model of sustained reading formalizes three lighting channels and generates six predictions all confirmed in 20 Monte Carlo simulations.
-
Multi-Agent Digital Twins for Strategic Decision-Making using Active Inference
Multi-agent active inference digital twins with contextual inference and streaming ML integration, illustrated via a Cournot competition socio-economic example.
-
Genetic Information as a "Chord" of Chemical Oscillations: Emergence of Catalyst-RNA Systems Driven by Superposed Rhythms
Simulations show that superposed Lotka-Volterra chemical oscillations can bias binary polymer elongation to produce catalytic loops and low-entropy informational sequences at higher rates than random selection.
-
Active inference as a unified model of collision avoidance behavior in human drivers
Active inference model unifies human collision avoidance by reproducing meta-analysis aggregates and simulator-specific effects on response timing, maneuver selection, and execution.