Enhanced representation learning with temporal coding in sparsely spiking neural networks.

Current representation learning methods in Spiking Neural Networks (SNNs) rely on rate-based encoding, resulting in high spike counts, increased energy consumption, and slower information transmission. In contrast, our proposed method, Weight-Temporally Coded Representation Learning (W-TCRL), utilizes temporally coded inputs, leading to lower spike counts and improved efficiency. To address the challenge of extracting representations from a temporal code with low reconstruction error, we introduce a novel Spike-Timing-Dependent Plasticity (STDP) rule. This rule enables stable learning of relative latencies within the synaptic weight distribution and is locally implemented in space and time, making it compatible with neuromorphic processors. We evaluate the performance of W-TCRL on the MNIST and natural image datasets for image reconstruction tasks. Our results demonstrate relative improvements of 53% for MNIST and 75% for natural images in terms of reconstruction error compared to the SNN state of the art. Additionally, our method achieves significantly higher sparsity, up to 900 times greater, when compared to related work. These findings emphasize the efficacy of W-TCRL in leveraging temporal coding for enhanced representation learning in Spiking Neural Networks.

Asymmetry in neural fields: a spatiotemporal encoding mechanism.

Neural field models have been successfully applied to model diverse brain mechanisms like visual attention, motor control, and memory. Most theoretical and modeling works have focused on the study of the dynamics of such systems under variations in neural connectivity, mainly symmetric connectivity among neurons. However, less attention has been given to the emerging properties of neuron populations when neural connectivity is asymmetric, although asymmetric activity propagation has been observed in cortical tissue. Here we explore the dynamics of neural fields with asymmetric connectivity and show, in the case of front propagation, that it can bias the population to follow a certain trajectory with higher activation. We find that asymmetry relates linearly to the input speed when the input is spatially localized, and this relation holds for different kernels and input shapes. To illustrate the behavior of asymmetric connectivity, we present an application: standard video sequences of human motion were encoded using the asymmetric neural field and compared to computer vision techniques. Overall, our results indicate that asymmetric neural fields are a competitive approach for spatiotemporal encoding with two main advantages: online classification and distributed operation.

The role of the asymptotic dynamics in the design of FPGA-based hardware implementations of gIF-type neural networks.

This paper presents a numerical analysis of the role of asymptotic dynamics in the design of hardware-based implementations of the generalised integrate-and-fire (gIF) neuron models. These proposed implementations are based on extensions of the discrete-time spiking neuron model, which was introduced by Soula et al., and have been implemented on Field Programmable Gate Array (FPGA) devices using fixed-point arithmetic. Mathematical studies conducted by Cessac have evidenced the existence of three main regimes (neural death, periodic and chaotic regimes) in the activity of such neuron models. These activity regimes are characterised in hardware by considering a precision analysis in the design of an architecture for an FPGA-based implementation. The proposed approach, although based on gIF neuron models and FPGA hardware, can be extended to more complex neuron models as well as to different in silico implementations.

Partial amplitude synchronization detection in brain signals using Bayesian Gaussian mixture models.

The present work investigates instantaneous synchronization in multivariate signals. It introduces a new method to detect subsets of synchronized time series that do not consider any baseline information. The method is based on a Bayesian Gaussian mixture model applied at each location of a time-frequency map. The work assesses the relevance of detected subsets by a stability measure. The application to Local Field Potentials measured during a visuo-motor experiment in monkeys reveals a subset of synchronized time series measured in the visual cortex.

Coupling BCM and neural fields for the emergence of self-organization consensus.

Human beings interact with the environment through different modalities, i.e. perceptions and actions, processed in the cortex by dedicated brain areas. These areas are self-organized, so that spatially close neurons are sensitive to close stimuli, providing generalization from previously learned examples. Although perceptive flows are picked up by different spatially separated sensors, their processings are not isolated. On the contrary, they are constantly interacting, as illustrated by the McGurk effect. When the auditory stimulus /ba/ and the /ga/ lip movement are presented simultaneously, people perceive a /da/, which does not correspond to any of the stimuli. Merging several stimuli into one multimodal perception reduces ambiguities and noises and is essential to interact with the environment. This article proposes a model for modality association, inspired by the biological properties of the cortex. The model consists of modality maps interacting through an associative map to raise a consistent multimodal perception of the environment. We propose the coupling of BCM learning rule and neural maps to obtain the decentralized and unsupervised self-organization of each modal map influenced by the multisensory context. We obtain local self-organization of modal maps with various inputs and discretizations.

Competition in high dimensional spaces using a sparse approximation of neural fields.

The Continuum Neural Field Theory implements competition within topologically organized neural networks with lateral inhibitory connections. However, due to the polynomial complexity of matrix-based implementations, updating dense representations of the activity becomes computationally intractable when an adaptive resolution or an arbitrary number of input dimensions is required. This paper proposes an alternative to self-organizing maps with a sparse implementation based on Gaussian mixture models, promoting a trade-off in redundancy for higher computational efficiency and alleviating constraints on the underlying substrate.This version reproduces the emergent attentional properties of the original equations, by directly applying them within a continuous approximation of a high dimensional neural field. The model is compatible with preprocessed sensory flows but can also be interfaced with artificial systems. This is particularly important for sensorimotor systems, where decisions and motor actions must be taken and updated in real-time. Preliminary tests are performed on a reactive color tracking application, using spatially distributed color features.

On-chip visual perception of motion: a bio-inspired connectionist model on FPGA.

Visual motion provides useful information to understand the dynamics of a scene to allow intelligent systems interact with their environment. Motion computation is usually restricted by real time requirements that need the design and implementation of specific hardware architectures. In this paper, the design of hardware architecture for a bio-inspired neural model for motion estimation is presented. The motion estimation is based on a strongly localized bio-inspired connectionist model with a particular adaptation of spatio-temporal Gabor-like filtering. The architecture is constituted by three main modules that perform spatial, temporal, and excitatory-inhibitory connectionist processing. The biomimetic architecture is modeled, simulated and validated in VHDL. The synthesis results on a Field Programmable Gate Array (FPGA) device show the potential achievement of real-time performance at an affordable silicon area.