Online Learning Of Neural Computations From Sparse Temporal Feedback
Dec 6, 2021
Speakers
Lukas Braun
Speaker · 0 followers
Tim P. Vogels
Speaker · 0 followers
About
Neuronal computations depend on synaptic connectivity and intrinsic electrophysiological properties. Synaptic connectivity determines which inputs from presynaptic neurons are integrated, while cellular properties determine how inputs are filtered over time. Unlike their biological counterparts, most computational approaches to learning in simulated neural networks are limited to changes in synaptic connectivity. However, if intrinsic parameters change, neural computations are altered drastically. Here, we include the parameters that determine the intrinsic properties, e.g., time constants and reset potential, into the learning paradigm. Using sparse feedback signals that indicate target spike times, and gradient-based parameter updates, we show that the intrinsic parameters can be learned along with the synaptic weights to produce specific input-output functions. Specifically, we use a teacher-student paradigm in which a randomly initialised leaky integrate-and-fire or resonate-and-fire neuron must recover the parameters of a teacher neuron. We show that complex temporal functions can be learned online and without backpropagation through time, relying on event-based updates only. Our results are a step towards online learning of neural computations from ungraded and unsigned sparse feedback signals with a biologically inspired learning mechanism.Neuronal computations depend on synaptic connectivity and intrinsic electrophysiological properties. Synaptic connectivity determines which inputs from presynaptic neurons are integrated, while cellular properties determine how inputs are filtered over time. Unlike their biological counterparts, most computational approaches to learning in simulated neural networks are limited to changes in synaptic connectivity. However, if intrinsic parameters change, neural computations are altered drasticall…
Organizer
NeurIPS 2021
Account · 1.9k followers
About NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Unlocking the Translational Potential of Resting-State Data
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Learning Abstractions for Robust and Tractable Planning
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Okachihuali: Strategies for new future building with AI art
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Truly Sparse Neural Network Training
Watch later
Xiao Zhou, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
PCA Retargeting: Encoding Linear Shape Models as Convolutional Mesh Autoencoders
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Fairness in Ranking under Uncertainty
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%