On the asymptotic equivalence between differential Hebbian and temporal difference learning

Authors:
Christoph Kolodziejski;Bernd Porr;Florentin Wörgötter
Affiliations:
Bernstein Center for Computational Neuroscience, University of Göttingen, Göttingen, Germany;Department of Electronics and Electrical Engineering, University of Glasgow, Glasgow, Scotland;Bernstein Center for Computational Neuroscience, University of Göttingen, Göttingen, Germany
Venue:
Neural Computation
Year:
2009

Citing 14
Cited 0

Differential Hebbian learning

AIP Conference Proceedings 151 on Neural Networks for Computing
Self-Organization in a Perceptual Network

Computer
Technical Note: \cal Q-Learning

Machine Learning
TD(λ) Converges with Probability 1

Machine Learning
Convergence Results for Single-Step On-PolicyReinforcement-Learning Algorithms

Machine Learning
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Learning to Predict by the Methods of Temporal Differences

Machine Learning
Spike-Timing-Dependent Hebbian Plasticity as Temporal Difference Learning

Neural Computation
Reinforcement Learning Through Modulation of Spike-Timing-Dependent Synaptic Plasticity

Neural Computation
Learning with “Relevance”: Using a Third Factor to Stabilize Hebbian Learning

Neural Computation
Mathematical properties of neuronal TD-rules and differential Hebbian learning: a comparison

Biological Cybernetics
An implementation of reinforcement learning based on spike timing dependent plasticity

Biological Cybernetics
A spiking neural network model of an actor-critic learning agent

Neural Computation
Experiments with infinite-horizon, policy-gradient estimation

Journal of Artificial Intelligence Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this theoretical contribution, we provide mathematical proof that two of the most important classes of network learning-correlation-based differential Hebbian learning and reward-based temporal difference learning--are asymptotically equivalent when timing the learning with a modulatory signal. This opens the opportunity to consistently reformulate most of the abstract reinforcement learning framework from a correlation-based perspective more closely related to the biophysics of neurons.