Unsupervised visual feature learning with spike-timing-dependent plasticity: How far are we from traditional feature learning approaches?. (September 2019)

• Record Type: Journal Article
• Authors: Falez, Pierre; Tirilly, Pierre; Bilasco, Ioan Marius; Devienne, Philippe; Boulet, Pierre
• Is Part Of: Pattern recognition. Volume 93(2019:Sep.)
• Page Start: 418
• Page End: 429
• Publication Date: 2019-09
• Subjects: Feature learning -- Unsupervised learning -- Spiking neural networks -- Spike-timing dependent plasticity -- Auto-encoders -- Image recognition; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2019.04.016
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We compare the performance of spiking neural networks (SNNs) with auto-encoders for visual feature learning. Results show that current SNNs are not competitive with autoencoders. From the analysis of the results, we identify some bottlenecks that should be addressed to make SNNs competitive with state-of-the-art systems. Abstract: Spiking neural networks (SNNs) equipped with latency coding and spike-timing dependent plasticity rules offer an alternative to solve the data and energy bottlenecks of standard computer vision approaches: they can learn visual features without supervision and can be implemented by ultra-low power hardware architectures. However, their performance in image classification has never been evaluated on recent image datasets. In this paper, we compare SNNs to auto-encoders on three visual recognition datasets, and extend the use of SNNs to color images. The analysis of the results helps us identify some bottlenecks of SNNs: the limits of on-center/off-center coding, especially for color images, and the ineffectiveness of current inhibition mechanisms. These issues should be addressed to build effective SNNs for image recognition.