Non-linear motor control by local learning in spiking neural networks

Aditya Gilra, Wulfram Gerstner
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:1773-1782, 2018.

Abstract

Learning weights in a spiking neural network with hidden neurons, using local, stable and online rules, to control non-linear body dynamics is an open problem. Here, we employ a supervised scheme, Feedback-based Online Local Learning Of Weights (FOLLOW), to train a heterogeneous network of spiking neurons with hidden layers, to control a two-link arm so as to reproduce a desired state trajectory. We show that the network learns an inverse model of the non-linear dynamics, i.e. it infers from state trajectory as input to the network, the continuous-time command that produced the trajectory. Connection weights are adjusted via a local plasticity rule that involves pre-synaptic firing and post-synaptic feedback of the error in the inferred command. We propose a network architecture, termed differential feedforward, and show that it gives a lower test error than other feedforward and recurrent architectures. We demonstrate the performance of the inverse model to control a two-link arm along a desired trajectory.

Cite this Paper


BibTeX
@InProceedings{pmlr-v80-gilra18a, title = {Non-linear motor control by local learning in spiking neural networks}, author = {Gilra, Aditya and Gerstner, Wulfram}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {1773--1782}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/gilra18a/gilra18a.pdf}, url = {http://proceedings.mlr.press/v80/gilra18a.html}, abstract = {Learning weights in a spiking neural network with hidden neurons, using local, stable and online rules, to control non-linear body dynamics is an open problem. Here, we employ a supervised scheme, Feedback-based Online Local Learning Of Weights (FOLLOW), to train a heterogeneous network of spiking neurons with hidden layers, to control a two-link arm so as to reproduce a desired state trajectory. We show that the network learns an inverse model of the non-linear dynamics, i.e. it infers from state trajectory as input to the network, the continuous-time command that produced the trajectory. Connection weights are adjusted via a local plasticity rule that involves pre-synaptic firing and post-synaptic feedback of the error in the inferred command. We propose a network architecture, termed differential feedforward, and show that it gives a lower test error than other feedforward and recurrent architectures. We demonstrate the performance of the inverse model to control a two-link arm along a desired trajectory.} }
Endnote
%0 Conference Paper %T Non-linear motor control by local learning in spiking neural networks %A Aditya Gilra %A Wulfram Gerstner %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-gilra18a %I PMLR %P 1773--1782 %U http://proceedings.mlr.press/v80/gilra18a.html %V 80 %X Learning weights in a spiking neural network with hidden neurons, using local, stable and online rules, to control non-linear body dynamics is an open problem. Here, we employ a supervised scheme, Feedback-based Online Local Learning Of Weights (FOLLOW), to train a heterogeneous network of spiking neurons with hidden layers, to control a two-link arm so as to reproduce a desired state trajectory. We show that the network learns an inverse model of the non-linear dynamics, i.e. it infers from state trajectory as input to the network, the continuous-time command that produced the trajectory. Connection weights are adjusted via a local plasticity rule that involves pre-synaptic firing and post-synaptic feedback of the error in the inferred command. We propose a network architecture, termed differential feedforward, and show that it gives a lower test error than other feedforward and recurrent architectures. We demonstrate the performance of the inverse model to control a two-link arm along a desired trajectory.
APA
Gilra, A. & Gerstner, W.. (2018). Non-linear motor control by local learning in spiking neural networks. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:1773-1782 Available from http://proceedings.mlr.press/v80/gilra18a.html.

Related Material