Nonlinear motor control by local learning in spiking neural networks
[edit]
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:17731782, 2018.
Abstract
Learning weights in a spiking neural network with hidden neurons, using local, stable and online rules, to control nonlinear body dynamics is an open problem. Here, we employ a supervised scheme, Feedbackbased Online Local Learning Of Weights (FOLLOW), to train a heterogeneous network of spiking neurons with hidden layers, to control a twolink arm so as to reproduce a desired state trajectory. We show that the network learns an inverse model of the nonlinear dynamics, i.e. it infers from state trajectory as input to the network, the continuoustime command that produced the trajectory. Connection weights are adjusted via a local plasticity rule that involves presynaptic firing and postsynaptic 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 twolink arm along a desired trajectory.
Related Material


