Cortical Microcircuits from a Generative Vision Model

TL;DR

This paper derives biologically plausible cortical microcircuit models from a generative vision model, linking anatomical features to computational roles in Bayesian inference for visual processing.

Contribution

It introduces a novel cortical circuit model grounded in a generative visual inference framework, validated with real-world tasks and aligned with anatomical constraints.

Findings

Functional roles assigned to cortical connections

Thalamic pathway's computational role identified

Model demonstrates efficient inference and generalization

Abstract

Understanding the information processing roles of cortical circuits is an outstanding problem in neuroscience and artificial intelligence. The theoretical setting of Bayesian inference has been suggested as a framework for understanding cortical computation. Based on a recently published generative model for visual inference (George et al., 2017), we derive a family of anatomically instantiated and functional cortical circuit models. In contrast to simplistic models of Bayesian inference, the underlying generative model's representational choices are validated with real-world tasks that required efficient inference and strong generalization. The cortical circuit model is derived by systematically comparing the computational requirements of this model with known anatomical constraints. The derived model suggests precise functional roles for the feedforward, feedback and lateral connections observed in different laminae and columns, and assigns a computational role for the path through the thalamus.