Optimizing Semantic Pointer Representations for Symbol-Like Processing in Spiking Neural Networks.
Ontology highlight
ABSTRACT:
SUBMITTER: Gosmann J
PROVIDER: S-EPMC4762696 | biostudies-literature | 2016
REPOSITORIES: biostudies-literature
Publications
Optimizing Semantic Pointer Representations for Symbol-Like Processing in Spiking Neural Networks.
Gosmann Jan J Eliasmith Chris C
PloS one 20160222 2
The Semantic Pointer Architecture (SPA) is a proposal of specifying the computations and architectural elements needed to account for cognitive functions. By means of the Neural Engineering Framework (NEF) this proposal can be realized in a spiking neural network. However, in any such network each SPA transformation will accumulate noise. By increasing the accuracy of common SPA operations, the overall network performance can be increased considerably. As well, the representations in such networ ...[more]
Similar Datasets
Project description:In the last few years, spiking neural networks (SNNs) have been demonstrated to perform on par with regular convolutional neural networks. Several works have proposed methods to convert a pre-trained CNN to a Spiking CNN without a significant sacrifice of performance. We demonstrate first that quantization-aware training of CNNs leads to better accuracy in SNNs. One of the benefits of converting CNNs to spiking CNNs is to leverage the sparse computation of SNNs and consequently perform equivalent computation at a lower energy consumption. Here we propose an optimization strategy to train efficient spiking networks with lower energy consumption, while maintaining similar accuracy levels. We demonstrate results on the MNIST-DVS and CIFAR-10 datasets.
| S-EPMC7339957 | biostudies-literature
Project description:Spiking neural networks (SNNs) are potentially highly efficient models for inference on fully parallel neuromorphic hardware, but existing training methods that convert conventional artificial neural networks (ANNs) into SNNs are unable to exploit these advantages. Although ANN-to-SNN conversion has achieved state-of-the-art accuracy for static image classification tasks, the following subtle but important difference in the way SNNs and ANNs integrate information over time makes the direct application of conversion techniques for sequence processing tasks challenging. Whereas all connections in SNNs have a certain propagation delay larger than zero, ANNs assign different roles to feed-forward connections, which immediately update all neurons within the same time step, and recurrent connections, which have to be rolled out in time and are typically assigned a delay of one time step. Here, we present a novel method to obtain highly accurate SNNs for sequence processing by modifying the ANN training before conversion, such that delays induced by ANN rollouts match the propagation delays in the targeted SNN implementation. Our method builds on the recently introduced framework of streaming rollouts, which aims for fully parallel model execution of ANNs and inherently allows for temporal integration by merging paths of different delays between input and output of the network. The resulting networks achieve state-of-the-art accuracy for multiple event-based benchmark datasets, including N-MNIST, CIFAR10-DVS, N-CARS, and DvsGesture, and through the use of spatio-temporal shortcut connections yield low-latency approximate network responses that improve over time as more of the input sequence is processed. In addition, our converted SNNs are consistently more energy-efficient than their corresponding ANNs.
| S-EPMC7214871 | biostudies-literature
Project description:Previous evidence from neuropsychological and neuroimaging studies suggests functional specialization for tools and related semantic knowledge in a left frontoparietal network. It is still debated whether these areas are involved in the representation of rudimentary movement-relevant knowledge regardless of semantic domains (animate vs. inanimate) or categories (tools vs. nontool objects). Here, we used fMRI to record brain activity while 13 volunteers performed two semantic judgment tasks on visually presented items from three different categories: animals, tools, and nontool objects. Participants had to judge two distinct semantic features: whether two items typically move in a similar way (e.g., a fan and a windmill move in circular motion) or whether they are usually found in the same environment (e.g., a seesaw and a swing are found in a playground). We investigated differences in overall activation (which areas are involved) as well as representational content (which information is encoded) across semantic features and categories. Results of voxel-wise mass univariate analysis showed that, regardless of semantic category, a dissociation emerges between processing information on prototypical location (involving the anterior temporal cortex and the angular gyrus) and movement (linked to left inferior parietal and frontal activation). Multivoxel pattern correlation analyses confirmed the representational segregation of networks encoding task- and category-related aspects of semantic processing. Taken together, these findings suggest that the left frontoparietal network is recruited to process movement properties of items (including both biological and nonbiological motion) regardless of their semantic category.
| S-EPMC7012372 | biostudies-literature
Project description:State-of-the-art computer vision systems use frame-based cameras that sample the visual scene as a series of high-resolution images. These are then processed using convolutional neural networks using neurons with continuous outputs. Biological vision systems use a quite different approach, where the eyes (cameras) sample the visual scene continuously, often with a non-uniform resolution, and generate neural spike events in response to changes in the scene. The resulting spatio-temporal patterns of events are then processed through networks of spiking neurons. Such event-based processing offers advantages in terms of focusing constrained resources on the most salient features of the perceived scene, and those advantages should also accrue to engineered vision systems based upon similar principles. Event-based vision sensors, and event-based processing exemplified by the SpiNNaker (Spiking Neural Network Architecture) machine, can be used to model the biological vision pathway at various levels of detail. Here we use this approach to explore structural synaptic plasticity as a possible mechanism whereby biological vision systems may learn the statistics of their inputs without supervision, pointing the way to engineered vision systems with similar online learning capabilities.
| S-EPMC6015816 | biostudies-literature
Project description:In computer simulations of spiking neural networks, often it is assumed that every two neurons of the network are connected by a probability of 2%, 20% of neurons are inhibitory and 80% are excitatory. These common values are based on experiments, observations, and trials and errors, but here, I take a different perspective, inspired by evolution, I systematically simulate many networks, each with a different set of parameters, and then I try to figure out what makes the common values desirable. I stimulate networks with pulses and then measure their: dynamic range, dominant frequency of population activities, total duration of activities, maximum rate of population and the occurrence time of maximum rate. The results are organized in phase diagram. This phase diagram gives an insight into the space of parameters - excitatory to inhibitory ratio, sparseness of connections and synaptic weights. This phase diagram can be used to decide the parameters of a model. The phase diagrams show that networks which are configured according to the common values, have a good dynamic range in response to an impulse and their dynamic range is robust in respect to synaptic weights, and for some synaptic weights they oscillates in ? or ? frequencies, independent of external stimuli.
| S-EPMC4349167 | biostudies-literature
Project description:Modeling and simulating the neural structures which make up our central neural system is instrumental for deciphering the computational neural cues beneath. Higher levels of biological plausibility usually impose higher levels of complexity in mathematical modeling, from neural to behavioral levels. This paper focuses on overcoming the simulation problems (accuracy and performance) derived from using higher levels of mathematical complexity at a neural level. This study proposes different techniques for simulating neural models that hold incremental levels of mathematical complexity: leaky integrate-and-fire (LIF), adaptive exponential integrate-and-fire (AdEx), and Hodgkin-Huxley (HH) neural models (ranged from low to high neural complexity). The studied techniques are classified into two main families depending on how the neural-model dynamic evaluation is computed: the event-driven or the time-driven families. Whilst event-driven techniques pre-compile and store the neural dynamics within look-up tables, time-driven techniques compute the neural dynamics iteratively during the simulation time. We propose two modifications for the event-driven family: a look-up table recombination to better cope with the incremental neural complexity together with a better handling of the synchronous input activity. Regarding the time-driven family, we propose a modification in computing the neural dynamics: the bi-fixed-step integration method. This method automatically adjusts the simulation step size to better cope with the stiffness of the neural model dynamics running in CPU platforms. One version of this method is also implemented for hybrid CPU-GPU platforms. Finally, we analyze how the performance and accuracy of these modifications evolve with increasing levels of neural complexity. We also demonstrate how the proposed modifications which constitute the main contribution of this study systematically outperform the traditional event- and time-driven techniques under increasing levels of neural complexity.
| S-EPMC5293783 | biostudies-literature
Project description:Somatosensation is composed of two distinct modalities: touch, arising from sensors in the skin, and proprioception, resulting primarily from sensors in the muscles, combined with these same cutaneous sensors. In contrast to the wealth of information about touch, we know quite less about the nature of the signals giving rise to proprioception at the cortical level. Likewise, while there is considerable interest in developing encoding models of touch-related neurons for application to brain machine interfaces, much less emphasis has been placed on an analogous proprioceptive interface. Here we investigate the use of Artificial Neural Networks (ANNs) to model the relationship between the firing rates of single neurons in area 2, a largely proprioceptive region of somatosensory cortex (S1) and several types of kinematic variables related to arm movement. To gain a better understanding of how these kinematic variables interact to create the proprioceptive responses recorded in our datasets, we train ANNs under different conditions, each involving a different set of input and output variables. We explore the kinematic variables that provide the best network performance, and find that the addition of information about joint angles and/or muscle lengths significantly improves the prediction of neural firing rates. Our results thus provide new insight regarding the complex representations of the limb motion in S1: that the firing rates of neurons in area 2 may be more closely related to the activity of peripheral sensors than it is to extrinsic hand position. In addition, we conduct numerical experiments to determine the sensitivity of ANN models to various choices of training design and hyper-parameters. Our results provide a baseline and new tools for future research that utilizes machine learning to better describe and understand the activity of neurons in S1.
| S-EPMC6449471 | biostudies-literature
Project description:Artificial Neural Networks (ANNs) are bio-inspired models of neural computation that have proven highly effective. Still, ANNs lack a natural notion of time, and neural units in ANNs exchange analog values in a frame-based manner, a computationally and energetically inefficient form of communication. This contrasts sharply with biological neurons that communicate sparingly and efficiently using isomorphic binary spikes. While Spiking Neural Networks (SNNs) can be constructed by replacing the units of an ANN with spiking neurons (Cao et al., 2015; Diehl et al., 2015) to obtain reasonable performance, these SNNs use Poisson spiking mechanisms with exceedingly high firing rates compared to their biological counterparts. Here we show how spiking neurons that employ a form of neural coding can be used to construct SNNs that match high-performance ANNs and match or exceed state-of-the-art in SNNs on important benchmarks, while requiring firing rates compatible with biological findings. For this, we use spike-based coding based on the firing rate limiting adaptation phenomenon observed in biological spiking neurons. This phenomenon can be captured in fast adapting spiking neuron models, for which we derive the effective transfer function. Neural units in ANNs trained with this transfer function can be substituted directly with adaptive spiking neurons, and the resulting Adaptive SNNs (AdSNNs) can carry out competitive classification in deep neural networks without further modifications. Adaptive spike-based coding additionally allows for the dynamic control of neural coding precision: we show empirically how a simple model of arousal in AdSNNs further halves the average required firing rate and this notion naturally extends to other forms of attention as studied in neuroscience. AdSNNs thus hold promise as a novel and sparsely active model for neural computation that naturally fits to temporally continuous and asynchronous applications.
| S-EPMC6332470 | biostudies-literature
Project description:The adaptive changes in synaptic efficacy that occur between spiking neurons have been demonstrated to play a critical role in learning for biological neural networks. Despite this source of inspiration, many learning focused applications using Spiking Neural Networks (SNNs) retain static synaptic connections, preventing additional learning after the initial training period. Here, we introduce a framework for simultaneously learning the underlying fixed-weights and the rules governing the dynamics of synaptic plasticity and neuromodulated synaptic plasticity in SNNs through gradient descent. We further demonstrate the capabilities of this framework on a series of challenging benchmarks, learning the parameters of several plasticity rules including BCM, Oja's, and their respective set of neuromodulatory variants. The experimental results display that SNNs augmented with differentiable plasticity are sufficient for solving a set of challenging temporal learning tasks that a traditional SNN fails to solve, even in the presence of significant noise. These networks are also shown to be capable of producing locomotion on a high-dimensional robotic learning task, where near-minimal degradation in performance is observed in the presence of novel conditions not seen during the initial training period.
| S-EPMC8493296 | biostudies-literature
Project description:Finding out the physical structure of neuronal circuits that governs neuronal responses is an important goal for brain research. With fast advances for large-scale recording techniques, identification of a neuronal circuit with multiple neurons and stages or layers becomes possible and highly demanding. Although methods for mapping the connection structure of circuits have been greatly developed in recent years, they are mostly limited to simple scenarios of a few neurons in a pairwise fashion; and dissecting dynamical circuits, particularly mapping out a complete functional circuit that converges to a single neuron, is still a challenging question. Here, we show that a recent method, termed spike-triggered non-negative matrix factorization (STNMF), can address these issues. By simulating different scenarios of spiking neural networks with various connections between neurons and stages, we demonstrate that STNMF is a persuasive method to dissect functional connections within a circuit. Using spiking activities recorded at neurons of the output layer, STNMF can obtain a complete circuit consisting of all cascade computational components of presynaptic neurons, as well as their spiking activities. For simulated simple and complex cells of the primary visual cortex, STNMF allows us to dissect the pathway of visual computation. Taken together, these results suggest that STNMF could provide a useful approach for investigating neuronal systems leveraging recorded functional neuronal activity.
| S-EPMC8659421 | biostudies-literature