Since the advent of mobile robots, obstacle detection has been a topic of great interest. It has also been a subject of study in neuroscience, where flying insects and bats could be considered two of the most interesting cases in terms of vision-based and sound-based mechanisms for obstacle detection, respectively. Currently, many studies focus on vision-based obstacle detection, but not many can be found regarding sound-based obstacle detection. This work focuses on the latter approach, which also makes use of a Spiking Neural Network to exploit the advantages of these architectures and achieve an approach closer to biology. The complete system was tested through a series of experiments that confirm the validity of the spiking architecture for obstacle detection. It is empirically demonstrated that, when the distance between the robot and the obstacle decreases, the output firing rate of the system increases in response as expected, and vice versa. Therefore, there is a direct relation between the two. Furthermore, there is a distance threshold between detectable and undetectable objects which is also empirically measured in this work. An in-depth study on how this system works at low level based on the Inter-Spike Interval concept was performed, which may be useful in the future development of applications based on spiking filters.

As neural interfaces become more advanced, there has been an increase in the volume and complexity of neural data recordings. These interfaces capture rich information about neural dynamics that call for efficient, real-time processing algorithms to spontaneously extract and interpret patterns of neural dynamics. Moreover, being able to do so in a fully unsupervised manner is critical as patterns in vast streams of neural data might not be easily identifiable by the human eye. Formal Deep Neural Networks (DNNs) have come a long way in performing pattern recognition tasks for various static and sequential pattern recognition applications. However, these networks usually require large labeled datasets for training and have high power consumption preventing their future embedding in active brain implants. An alternative aimed at addressing these issues are Spiking Neural Networks (SNNs) which are neuromorphic and use more biologically plausible neurons with evolving membrane potentials. In this context, we introduce here a frugal single-layer SNN designed for fully unsupervised identification and classification of multivariate temporal patterns in continuous data with a sequential approach. We show that, with only a handful number of neurons, this strategy is efficient to recognize highly overlapping multivariate temporal patterns, first on simulated data, and then on Mel Cepstral representations of speech sounds and finally on multichannel neural data. This approach relies on several biologically inspired plasticity rules, including Spike-timing-dependent plasticity (STDP), Short-term plasticity (STP) and intrinsic plasticity (IP). These results pave the way towards highly frugal SNNs for fully unsupervised and online-compatible learning of complex multivariate temporal patterns for future embedding in dedicated very-low power hardware.

This study proposes a framework that extends existing time-coding time-to-first-spike spiking neural network (SNN) models to allow processing information changing over time. We explain spike propagation through a model with multiple input and output spikes at each neuron, as well as design training rules for end-to-end backpropagation. This strategy enables us to process information changing over time. The model is trained and evaluated on a Twitter bot detection task where the time of events (tweets and retweets) is the primary carrier of information. This task was chosen to evaluate how the proposed SNN deals with spike train data composed of hundreds of events occurring at timescales differing by almost five orders of magnitude. The impact of various parameters on model properties, performance and training-time stability is analyzed.

Spiking neural networks (SNNs) inherently rely on the timing of signals for representing and processing information. Transmission delays play an important role in shaping these temporal characteristics. Recent work has demonstrated the substantial advantages of learning these delays along with synaptic weights, both in terms of accuracy and memory efficiency. However, these approaches suffer from drawbacks in terms of precision and efficiency, as they operate in discrete time and with approximate gradients, while also requiring membrane potential recordings for calculating parameter updates. To alleviate these issues, we propose an analytical approach for calculating exact loss gradients with respect to both synaptic weights and delays in an event-based fashion. The inclusion of delays emerges naturally within our proposed formalism, enriching the model's search space with a temporal dimension. Our algorithm is purely based on the timing of individual spikes and does not require access to other variables such as membrane potentials. We explicitly compare the impact on accuracy and parameter efficiency of different types of delays - axonal, dendritic and synaptic. Furthermore, while previous work on learnable delays in SNNs has been mostly confined to software simulations, we demonstrate the functionality and benefits of our approach on the BrainScaleS-2 neuromorphic platform.

We propose a new interpretation of spiking neurons as Bayesian integra- tors accumulating evidence over time about events in the external world or the body, and communicating to other neurons their certainties about these events. In this model, spikes signal the occurrence of new infor- mation, i.e. what cannot be predicted from the past activity. As a result, firing statistics are close to Poisson, albeit providing a deterministic rep- resentation of probabilities. We proceed to develop a theory of Bayesian inference in spiking neural networks, recurrent interactions implement- ing a variant of belief propagation. Many perceptual and motor tasks performed by the central nervous system are probabilis- tic, and can be described in a Bayesian framework [4, 3].

As animals interact with their environments, they must constantly update estimates about their states. Bayesian models combine prior probabil- ities, a dynamical model and sensory evidence to update estimates op- timally. These models are consistent with the results of many diverse psychophysical studies. However, little is known about the neural rep- resentation and manipulation of such Bayesian information, particularly in populations of spiking neurons. We consider this issue, suggesting a model based on standard neural architecture and activations.

There's a supervised learning algorithm for SNN that enables a single neuron to learn spike pattern associations of input-output spike sequences at the precise times of spikes. This algorithm is termed SPAN(Spike Pattern Association Neuron). Anyone can build SNN to associate the input to output temporal patterns of desired spike sequences using this SPAN neuron. Here the input, output, and desired spike trains are transformed into analog signals by convolving the spikes with a kernel function. This transformation simplifies the computation of the error signal and, therefore, allows the application of gradient descent to optimize the synaptic weights.

Spiking Neural Networks (SNNs) (Gerstner and Kistler, 2002) have been shown to be computationally more powerful compared to traditional Artificial Neural Networks (Maass, 1997), even on the level of single neurons with single output spikes (Rubin et al., 2010). Though the computational power of SNNs has been demonstrated, practical applications are limited by their complexity. Large models with many parameters and high precision requirements are expensive to simulate and train, thus cannot meet the demands of real-time applications (Querlioz et al., 2013; Diehl and Cook, 2015; Balaji et al., 2020). While there are recent efforts (Yu et al., 2013b; Xu et al., 2018; Yu et al., 2019; Cheng et al., 2020; Li and Yu, 2020) to design smaller architectures which maintain competitive accuracy, it remains a significant challenge to analytically determine what SNN architecture, connectivity, or size are sufficient to enable robust capacity, even on elementary problems. In order to better understand the computational properties of SNNs, more efficient learning methods are required to enable further explorations of the capabilities of individual nodes in a network.