A biological gradient descent for prediction through a combination of STDP and homeostatic plasticity

DOI10.1162/NECO_A_00512zbMATH Open1415.92017arXiv1206.4812WikidataQ44739266 ScholiaQ44739266MaRDI QIDQ5378278FDOQ5378278

Authors: Gilles Wainrib, Mathieu N. Galtier 

Publication date: 12 June 2019

Published in: Neural Computation (Search for Journal in Brave)

Abstract: Identifying, formalizing and combining biological mechanisms which implement known brain functions, such as prediction, is a main aspect of current research in theoretical neuroscience. In this letter, the mechanisms of Spike Timing Dependent Plasticity (STDP) and homeostatic plasticity, combined in an original mathematical formalism, are shown to shape recurrent neural networks into predictors. Following a rigorous mathematical treatment, we prove that they implement the online gradient descent of a distance between the network activity and its stimuli. The convergence to an equilibrium, where the network can spontaneously reproduce or predict its stimuli, does not suffer from bifurcation issues usually encountered in learning in recurrent neural networks.

Full work available at URL: https://arxiv.org/abs/1206.4812

Recommendations

• Evaluating the extent to which homeostatic plasticity learns to compute prediction errors in unstructured neuronal networks
• Adaptive Synchronization of Activities in a Recurrent Network
• Spike-timing-dependent Hebbian plasticity as temporal difference learning
• Predictive learning in rate-coded neuronal networks: a theoretical approach towards classical conditioning
• Fading memory and time series prediction in recurrent networks with different forms of plasticity

zbMATH Keywords

predictionspike-timing-dependent plasticityrecurrent neural networkshomeostatic plasticity

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Neural networks for/in biological studies, artificial life and related topics (92B20)

Cites Work

• Nonlinear systems
• Averaging methods in nonlinear dynamical systems
• Spiking Neuron Models
• Bayesian Spiking Neurons I: Inference
• Relating STDP to BCM
• A simplified neuron model as a principal component analyzer
• Spike-timing-dependent Hebbian plasticity as temporal difference learning
• Recognizing recurrent neural networks (rRNN): Bayesian inference for recurrent neural net\-works
• Inverse Problems in Neural Field Theory
• Reservoir computing approaches to recurrent neural network training
• Why spikes? Hebbian learning and retrieval of time-resolved excitation patterns
• Fading memory and time series prediction in recurrent networks with different forms of plasticity
• Multiscale analysis of slow-fast neuronal learning models with noise

Cited In (7)

• A local echo state property through the largest Lyapunov exponent
• Dynamic branching in a neural network model for probabilistic prediction of sequences
• Distributed synaptic weights in a LIF neural network and learning rules
• Relative entropy minimizing noisy non-linear neural network to approximate stochastic processes
• Evaluating the extent to which homeostatic plasticity learns to compute prediction errors in unstructured neuronal networks
• Slow feature analysis with spiking neurons and its application to audio stimuli
• Fading memory and time series prediction in recurrent networks with different forms of plasticity

This page was built for publication: A biological gradient descent for prediction through a combination of STDP and homeostatic plasticity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5378278)