Biologically Plausible Sequence Learning with Spiking Neural Networks
This work addresses sequence learning for biologically inspired neural networks, offering a biologically plausible model with local learning rules, though it appears incremental in extending existing models like the Hopfield network.
The authors tackled the problem of sequence learning in spiking neural networks by proposing a novel continuous-time model called the McCulloch-Pitts network (MPN), which robustly memorizes multiple spatiotemporal patterns of binary vectors and efficiently learns sequences of binary pictures and generative models for neural spike-train data.
Motivated by the celebrated discrete-time model of nervous activity outlined by McCulloch and Pitts in 1943, we propose a novel continuous-time model, the McCulloch-Pitts network (MPN), for sequence learning in spiking neural networks. Our model has a local learning rule, such that the synaptic weight updates depend only on the information directly accessible by the synapse. By exploiting asymmetry in the connections between binary neurons, we show that MPN can be trained to robustly memorize multiple spatiotemporal patterns of binary vectors, generalizing the ability of the symmetric Hopfield network to memorize static spatial patterns. In addition, we demonstrate that the model can efficiently learn sequences of binary pictures as well as generative models for experimental neural spike-train data. Our learning rule is consistent with spike-timing-dependent plasticity (STDP), thus providing a theoretical ground for the systematic design of biologically inspired networks with large and robust long-range sequence storage capacity.