Abstract:Cortical populations of neurons develop sparse representations adapted to the statistics of the environment. To learn efficient population codes, synaptic plasticity mechanisms must differentiate relevant latent features from spurious input correlations, which are omnipresent in cortical networks. Here, we develop a theory for sparse coding and synaptic plasticity that is invariant to second-order correlations in the input. Going beyond classical Hebbian learning, our learning objective explains the functional form of observed excitatory plasticity mechanisms, showing how Hebbian long-term depression (LTD) cancels the sensitivity to second-order correlations so that receptive fields become aligned with features hidden in higher-order statistics. Invariance to second-order correlations enhances the versatility of biologically realistic learning models, supporting optimal decoding from noisy inputs and sparse population coding from spatially correlated stimuli. In a spiking model with triplet spike-timing-dependent plasticity (STDP), we show that individual neurons can learn localized oriented receptive fields, circumventing the need for input preprocessing, such as whitening, or population-level lateral inhibition. The theory advances our understanding of local unsupervised learning in cortical circuits, offers new interpretations of the Bienenstock-Cooper-Munro and triplet STDP models, and assigns a specific functional role to synaptic LTD mechanisms in pyramidal neurons. To understand how our brains carve out meaningful stimuli from a sea of sensory information, experimentalists often focus on individual neurons and their receptive fields; i.e., the input patterns that best stimulate the neurons. The receptive field of a neuron adapts to the statistics of the environment through plastic changes in the structure and strength of its incoming connections. Prevailing models of this adaptation, such as Sparse Coding and BCM theory, suggest that these connections are adjusted in a self-organized, unsupervised manner such that neurons can extract relatively rare, but complex patterns from the sensory input stream. Our theory extends these models, demonstrating that certain variants of the BCM model are capable of selectively ignoring pervasive second-order correlations and homing in on the pertinent, sparse signals in the stream of sensory stimuli. We have identified a carefully calibrated interaction between two aspects of synaptic plasticity: linear long-term depression (LTD) and nonlinear long-term potentiation (LTP). Such a balance enables a learning strategy that is sensitive to the complex statistical properties (higher-order correlations) of sensory inputs. In contrast to earlier approaches, so-called pre-whitening to remove second-order correlations is not necessary. Our findings provide not only a new framework for understanding the diverse roles of synaptic plasticity but also offer insights that may inform the development of artificial neural networks.

Adaptive compressed sensing - a new class of self-organizing coding models for neuroscience

Sparsity and Compressed Coding in Sensory Systems

Emergence of Sparse Coding, Balance and Decorrelation from a Biologically-Grounded Spiking Neural Network Model of Learning in the Primary Visual Cortex

A Sparse Coding Model with Synaptically Local Plasticity and Spiking Neurons Can Account for the Diverse Shapes of V1 Simple Cell Receptive Fields

Emergence of simple-cell receptive field properties by learning a sparse code for natural images

A Brain-Inspired Spiking Neural Network Model with Temporal Encoding and Learning

Sparse deep predictive coding captures contour integration capabilities of the early visual system

Cooperative coding of continuous variables in networks with sparsity constraint

Robust Transcoding Sensory Information with Neural Spikes

Sparse, Geometric Autoencoder Models of V1

The Relative Advantages of Sparse Versus Distributed Encoding for Associative Neuronal Networks in the Brain

Predictive coding with spiking neurons and feedforward gist signaling

Sparse coding with a somato-dendritic rule

Efficient Image Processing Via Compressive Sensing of Integrate-and-fire Neuronal Network Dynamics

Neural Elements for Predictive Coding

Efficient Visual Coding: From Retina To V2

Rapid learning of neural circuitry from holographic ensemble stimulation enabled by model-based compressed sensing

Learning what matters: Synaptic plasticity with invariance to second-order input correlations

Cortical Motion Perception Emerges from Dimensionality Reduction with Evolved Spike-Timing-Dependent Plasticity Rules

Adaptive coding efficiency in recurrent cortical circuits via gain control