Homotopy Theoretic and Categorical Models of Neural Information Networks
arXiv:2006.15136 · doi:10.46298/compositionality-6-4
Abstract
In this paper we develop a novel mathematical formalism for the modeling of neural information networks endowed with additional structure in the form of assignments of resources, either computational or metabolic or informational. The starting point for this construction is the notion of summing functors and of Segal's Gamma-spaces in homotopy theory. The main results in this paper include functorial assignments of concurrent/distributed computing architectures and associated binary codes to networks and their subsystems, a categorical form of the Hopfield network dynamics, which recovers the usual Hopfield equations when applied to a suitable category of weighted codes, a functorial assignment to networks of corresponding information structures and information cohomology, and a cohomological version of integrated information.
arXiv commentpublished version: 86 pages LaTeX
References in corpus (16)
- Clique topology reveals intrinsic geometric structure in neural correlations
- A mathematical theory of resources
- A unified framework for information integration based on information geometry
- A synthetic approach to Markov kernels, conditional independence and theorems on sufficient statistics
- A Characterization of Entropy in Terms of Information Loss
- Homotopy invariants of higher dimensional categories and concurrency in computer science
- The topology of the directed clique complex as a network invariant
- Bimonoidal Structure of Probability Monads
- Path homologies of deep feedforward networks
- Gamma Spaces and Information
- A model category for local po-spaces
- Robust motifs of threshold-linear networks
- Simplicial models for concurrency
- Stochastic order on metric spaces and the ordered Kantorovich monad
- The neural ring: using algebraic geometry to analyze neural codes
- Finite Automata with Time-Delay Blocks (Extended Version)