Cortical Microcircuits from a Generative Vision Model
This work addresses the problem of interpreting cortical microcircuits for neuroscience and AI researchers, but it appears incremental as it builds on an existing generative model.
The authors tackled the problem of understanding cortical circuit functions by deriving a family of anatomically instantiated cortical circuit models from a validated generative vision model, suggesting precise functional roles for feedforward, feedback, lateral connections, and thalamic pathways.
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.