Application of a Brain-Inspired Spiking Neural Network Architecture to Odor Data Classification.

Anup Vanarse,Adam Osseiran,Alexander Rassau,Nikola Kasabov,Josafath Israel Espinosa-Ramos

doi:10.3390/s20102756

Abstract

Existing methods in neuromorphic olfaction mainly focus on implementing the data transformation based on the neurobiological architecture of the olfactory pathway. While the transformation is pivotal for the sparse spike-based representation of odor data, classification techniques based on the bio-computations of the higher brain areas, which process the spiking data for identification of odor, remain largely unexplored. This paper argues that brain-inspired spiking neural networks constitute a promising approach for the next generation of machine intelligence for odor data processing. Inspired by principles of brain information processing, here we propose the first spiking neural network method and associated deep machine learning system for classification of odor data. The paper demonstrates that the proposed approach has several advantages when compared to the current state-of-the-art methods. Based on results obtained using a benchmark dataset, the model achieved a high classification accuracy for a large number of odors and has the capacity for incremental learning on new data. The paper explores different spike encoding algorithms and finds that the most suitable for the task is the step-wise encoding function. Further directions in the brain-inspired study of odor machine classification include investigation of more biologically plausible algorithms for mapping, learning, and interpretation of odor data along with the realization of these algorithms on some highly parallel and low power consuming neuromorphic hardware devices for real-world applications.

Highlights

Biological sensory architectures found in nature exhibit remarkable computational abilities and have the capacity to perform efficiently and accurately, even under noisy conditions [1]
The development of bioinspired learning methods such as spike-timing-dependent plasticity (STDP) and advancements in the utilization of spiking neural networks (SNN) for classification of odors based on temporal spiking information, reinforced the applicability of this approach for the development of robust and real-time electronic nose systems [3,6]
Based on the analysis provided in [37], Ben’s spiker algorithm (BSA) encoding may not be the ideal candidate for encoding rapidly changing signals, which is the case for electronic nose systems

Summary

Introduction

Biological sensory architectures found in nature exhibit remarkable computational abilities and have the capacity to perform efficiently and accurately, even under noisy conditions [1]. Pursuing the idea of replicating the same efficient style of computation, foundational research [2] by Persaud and Dodd aimed to develop an artificial olfactory system based on the functional blocks of the biological olfactory pathway. While this study introduced the notion of using a sensor array as the sensing front-end and established a general architecture for electronic nose (e-nose) systems, the implementation of conventional statistical methods to process multivariate time-series sensing data imposed limitations due to substantial computational latency, high power requirements, and poor classification performance and reliability [3]. The development of bioinspired learning methods such as spike-timing-dependent plasticity (STDP) and advancements in the utilization of spiking neural networks (SNN) for classification of odors based on temporal spiking information, reinforced the applicability of this approach for the development of robust and real-time electronic nose systems [3,6]

Methods

Results

Conclusion