Spiking neural networks, thanks to their sensitivity to the timing of the inputs, are a promising tool for unsupervised processing of spatio-temporal data. However, they do not perform as well as the traditional machine learning approaches and their real-world applications are still limited. Various supervised and reinforcement learning methods for optimising spiking neural networks have been proposed, but more recently the evolutionary approach regained attention as a tool for training neural networks. Here, we describe a simple evolutionary approach for optimising spiking neural networks. This is the first published use of evolutionary algorithm to develop hyperparameters for fully unsupervised spike-timing-dependent learning for pattern clustering using spiking neural networks. Our results show that combining evolution and unsupervised learning leads to faster convergence on the optimal solutions, better stability of fit solutions and higher fitness of the whole population than using each approach separately.

Neuromodulation is a pervasive biological process impacting neural activity at many scales. Changes in the concentration of a single neuromodulator can drastically alter the dynamics of a circuit. Nevertheless, how circuits can be both sensitive to the effects of neuromodulators, yet maintain stable behaviors in the face of constantly changing concentrations of them, is still poorly understood. Past work addressing this has focused on isolated circuits or individual neurons. In this paper, we study the effects of neuromodulation in the context of a complete brain-body-environment model. We use a genetic algorithm to find configurations of a dynamical neural network able to walk with and without the presence of an extrinsic neuromodulatory signal. We analyze, in some detail, networks, which break and cope under the effects of neuromodulation. We identify common stability mechanisms among successful networks, which correspond to previously proposed ideas. In addition, results indicate that proprioceptive feedback provides a stability mechanism for coping with neuromodulation that has not previously been considered in the literature.

Grid cells in the hippocampal formation are a valuable system to study both for neuroscientists and for neural network researchers, as these neurons present both a window into higher-level cognitive processes such as navigation, as well as inspiration for how to build artificial neural navigation systems. Grid cells are believed to represent an animal’s coordinates in two-dimensional space in a general fashion, useable for geometric computations by downstream neural networks, and earlier neural models have indeed shown how grid cells can be decoded for navigational purposes. However, accumulating evidence shows that grid cells are not as stable as assumed by models, but that they exhibit various geometric distortions depending on time and place. This presents a challenge to grid cell decoding models, which mainly separate into “nested” and “combinatorial” ones. Here we present a new and simplified version of a nested grid cell decoder, demonstrate that this decoder can cope with distortions, and show how this relates to a fundamental property of nested grid cell decoding. By providing positive proof that a nested decoder can navigate with distorted grid cells, we hope to inspire further neuroscientific investigation into the biological plausibility of different models for grid cell-based navigation.