Bayesian reconstruction of memories stored in neural networks from their connectivity

• Published in: PLoS computational biology Vol. 19; no. 1; p. e1010813
• Main Authors: Goldt, Sebastian, Krzakala, Florent, Zdeborová, Lenka, Brunel, Nicolas
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 01.01.2023; Public Library of Science (PLoS)
• Subjects: Algorithms; Bayes Theorem; Bayesian analysis; Bayesian statistical decision theory; Biology and Life Sciences; Book publishing; Circuit diagrams; Computer and Information Sciences; Information storage; Mathematical models; Medicine and Health Sciences; Memory; Models, Neurological; Neural circuitry; Neural networks; Neural Networks, Computer; Neurons; Neurons - physiology; Phase transitions; Physical Sciences; Physiological aspects; Questions; Random variables; Research and Analysis Methods; Statistical analysis; Statistical inference; Wiring; Italy; Iran; United Kingdom
• ISSN: 1553-7358; 1553-734X; 1553-7358
• DOI: 10.1371/journal.pcbi.1010813

The advent of comprehensive synaptic wiring diagrams of large neural circuits has created the field of connectomics and given rise to a number of open research questions. One such question is whether it is possible to reconstruct the information stored in a recurrent network of neurons, given its synaptic connectivity matrix. Here, we address this question by determining when solving such an inference problem is theoretically possible in specific attractor network models and by providing a practical algorithm to do so. The algorithm builds on ideas from statistical physics to perform approximate Bayesian inference and is amenable to exact analysis. We study its performance on three different models, compare the algorithm to standard algorithms such as PCA, and explore the limitations of reconstructing stored patterns from synaptic connectivity.