Abstract:Over the past several decades, backpropagation (BP) has played a critical role in the advancement of machine learning and remains a core method in numerous computational applications. It is also utilized extensively in comparative studies of biological and artificial neural network representations. Despite its widespread use, the implementation of BP in the brain remains elusive, and its biological plausibility is often questioned due to inherent issues such as the need for symmetry of weights between forward and backward connections, and the requirement of distinct forward and backward phases of computation. Here, we introduce a novel neuroplasticity rule that offers a potential mechanism for implementing BP in the brain. Similar in general form to the classical Hebbian rule, this rule is based on the core principles of maintaining the balance of excitatory and inhibitory inputs as well as on retrograde signaling, and operates over three progressively slower timescales: neural firing, retrograde signaling, and neural plasticity. We hypothesize that each neuron possesses an internal state, termed credit, in addition to its firing rate. After achieving equilibrium in firing rates, neurons receive credits based on their contribution to the E-I balance of postsynaptic neurons through retrograde signaling. As the network's credit distribution stabilizes, connections from those presynaptic neurons are strengthened that significantly contribute to the balance of postsynaptic neurons. We demonstrate mathematically that our learning rule precisely replicates BP in layered neural networks without any approximations. Simulations on artificial neural networks reveal that this rule induces varying community structures in networks, depending on the learning rate. This simple theoretical framework presents a biologically plausible implementation of BP, with testable assumptions and predictions that may be evaluated through biological experiments.

Activation Relaxation: A Local Dynamical Approximation to Backpropagation in the Brain

Investigating the Scalability and Biological Plausibility of the Activation Relaxation Algorithm

Backpropagation through space, time, and the brain

Dendritic cortical microcircuits approximate the backpropagation algorithm

Approximating Back-propagation for a Biologically Plausible Local Learning Rule in Spiking Neural Networks

Generalization of Equilibrium Propagation to Vector Field Dynamics

Activation Learning by Local Competitions

Contribute to balance, wire in accordance: Emergence of backpropagation from a simple, bio-plausible neuroplasticity rule

Towards Biologically Plausible Computing: A Comprehensive Comparison

A More Biologically Plausible Local Learning Rule for ANNs

Backprop Diffusion is Biologically Plausible

Learning efficient backprojections across cortical hierarchies in real time

Hebbian Deep Learning Without Feedback

A Theory of Local Learning, the Learning Channel, and the Optimality of Backpropagation

Error Forward-Propagation: Reusing Feedforward Connections to Propagate Errors in Deep Learning

Dendritic error backpropagation in deep cortical microcircuits

A biologically plausible neural network for local supervision in cortical microcircuits

Evolutionary algorithms as an alternative to backpropagation for supervised training of Biophysical Neural Networks and Neural ODEs

A New Backpropagation Algorithm without Gradient Descent

Single-phase deep learning in cortico-cortical networks