Fluctuation-Dissipation Theorem and Models of Learning
From MaRDI portal
Abstract: Advances in statistical learning theory have resulted in a multitude of different designs of learning machines. But which ones are implemented by brains and other biological information processors? We analyze how various abstract Bayesian learners perform on different data and argue that it is difficult to determine which learning-theoretic computation is performed by a particular organism using just its performance in learning a stationary target (learning curve). Basing on the fluctuation-dissipation relation in statistical physics, we then discuss a different experimental setup that might be able to solve the problem.
Recommendations
Cites work
- Bayesian Computation in Recurrent Neural Circuits
- Density estimation by stochastic complexity
- Estimating the dimension of a model
- Field Theories for Learning Probability Distributions
- I-divergence geometry of probability distributions and minimization problems
- Information-theoretic asymptotics of Bayes methods
- On stochastic complexity and nonparametric density estimation
- On the mathematical foundations of learning
- Predictability, complexity, and learning
- Present Position and Potential Developments: Some Personal Views: Statistical Theory: The Prequential Approach
- Statistical Inference, Occam's Razor, and Statistical Mechanics on the Space of Probability Distributions
Cited in
(3)
This page was built for publication: Fluctuation-Dissipation Theorem and Models of Learning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5706657)
