Y Bengio, A Fischer - arXiv preprint arXiv:1510.02777, 2015 - arxiv.org
We show that Langevin MCMC inference in an energy-based model with latent variables has the property that the early steps of inference, starting from a stationary point, correspond …
B Scellier, Y Bengio - Frontiers in computational neuroscience, 2017 - frontiersin.org
We introduce Equilibrium Propagation, a learning framework for energy-based models. It involves only one kind of neural computation, performed in both the first phase (when the …
Ongoing advances in experimental technique are making commonplace simultaneous recordings of the activity of tens to hundreds of cortical neurons at high temporal resolution …
How the brain performs credit assignment is a fundamental unsolved problem in neuroscience. Manybiologically plausible'algorithms have been proposed, which compute …
Abstract Models of sensory processing and learning in the cortex need to efficiently assign credit to synapses in all areas. In deep learning, a known solution is error backpropagation …
Experimental evidence indicates that synaptic modification depends on the timing relationship between the presynaptic inputs and the output spikes that they generate. In this …
Deep learning has seen remarkable developments over the last years, many of them inspired by neuroscience. However, the main learning mechanism behind these advances …
This review article summarises recently proposed theories on how neural circuits in the brain could approximate the error back-propagation algorithm used by artificial neural …
S Linderman, CH Stock… - Advances in neural …, 2014 - proceedings.neurips.cc
Learning and memory in the brain are implemented by complex, time-varying changes in neural circuitry. The computational rules according to which synaptic weights change over …