Exploiting Multiple Timescales in Hierarchical Echo State Networks

Abstract

Echo state networks (ESNs) are a powerful form of reservoir computing that only require training of linear output weights while the internal reservoir is formed of fixed randomly connected neurons. With a correctly scaled connectivity matrix, the neurons’ activity exhibits the echo-state property and responds to the input dynamics with certain timescales. Tuning the timescales of the network can be necessary for treating certain tasks, and some environments require multiple timescales for an efficient representation. Here we explore the timescales in hierarchical ESNs, where the reservoir is partitioned into two smaller linked reservoirs with distinct properties. Over three different tasks (NARMA10, a reconstruction task in a volatile environment, and psMNIST), we show that by selecting the hyper-parameters of each partition such that they focus on different timescales, we achieve a significant performance improvement over a single ESN. Through a linear analysis, and under the assumption that the timescales of the first partition are much shorter than the second’s (typically corresponding to optimal operating conditions), we interpret the feedforward coupling of the partitions in terms of an effective representation of the input signal, provided by the first partition to the second, whereby the instantaneous input signal is expanded into a weighted combination of its time derivatives. Furthermore, we propose a data-driven approach to optimise the hyper-parameters through a gradient descent optimisation method that is an online approximation of backpropagation through time. We demonstrate the application of the online learning rule across all the tasks considered.

Signal neutrality, scalar property, and collapsing boundaries as consequences of a learned multi-timescale strategy

“…Timescales have been implicit in the reinforcement learning framework, in the context of propagating information about the success (or failure) of the task in cases where reward is not immediate, see for instance eligibility traces [43,59]. This is not entirely the same as the concept of time scales in this model, where the emphasis is on acquiring and retaining sensory information from the environment, not unlike what happens in the field of Reservoir Computing [60].…”

Voltage-controlled superparamagnetic ensembles for low-power reservoir computing

“…The complexity of the maps, while not discussed in further detail here, may be due in part to coupling between input strength and internal timescale (see supplementary). The goal of the chaotic time series task (NARMA10) is to predict the response to white noise (random numbers drawn from [0, 0.5]) of a discrete-time tenth order nonlinear auto-regressive moving average 39,40 :…”

Compensatory variability in network parameters enhances memory performance in theDrosophilamushroom body

“…Yet even in populations that ostensibly encode the same kind of stimulus, like olfactory mitral cells, heterogeneity of neuronal excitability can increase the information content of their population activity [15][16][17]. In addition, heterogeneity in neuronal time scales can improve learning in neural networks [18,19]. In what contexts and in what senses might the opposite be true, i.e., when does neuronal similarity provide computational benefits over neuronal diversity?…”