Topic: Gaussian model

Neural circuits must perform probabilistic computations to efficiently process noisy sensory information. Sampling codes propose that single neuron variability is a signature of probabilistic computation, and corresponds to sampling the space of possible solutions in proportion to their posterior probability. Under this hypothesis, the neural code defines how variables inferred by the network are represented in single neurons. Compared to other proposed probabilistic codes, the capacity of sampling codes scales with the number of neurons, but their convergence speed scales poorly with dimensio

Cortical neurons exhibit a high degree of trial-to-trial variability in response to repeated presentations of the same stimulus. We propose that this variability serves a useful computational purpose: allowing the brain to generalize from a small number of examples. To formalize this theory, we consider a Gaussian model of few-shot classification, wherein we observe a few samples of each class, and each observation has trial-to-trial variability. Our model leads to three predictions about the optimal neural response variability that minimizes generalization error: (i) the optimal variability m