Skip to Main content Skip to Navigation
Conference papers

A homeostatic gain control mechanism to improve event-driven object recognition

Abstract : We propose a neuromimetic architecture able to perform pattern recognition. To achieve this, we extended the existing event-based algorithm from [1] which introduced novel spatio-temporal features: time surfaces. Built from asynchronous events acquired by a neuromorphic camera, these time surfaces allow to code the local dynamics of a visual scene and create an efficient hierarchical event-based pattern recognition architecture. Inspired by biological findings and the efficient coding hypothesis, our main contribution is to integrate homeostatic regulation into the Hebbian learning rule. Indeed, in order to be optimally informative, average neural activity within a layer should be equally balanced across neurons. We used that principle to regularize neurons within the same layer by setting a gain depending on their past activity and such that they emit spikes with balanced firing rates. The efficiency of this technique was first demonstrated through a robust improvement in spatiotemporal patterns which were learnt during the training phase. In order to compare with state-of-the-art methods, we replicated past results on the same dataset as [1] and extended results in this study to the widely used N-MNIST dataset [2]. We expect to extend this fully event-driven approach to more naturalistic tasks, notably for ultra-fast object categorization.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03336554
Contributor : Laurent Perrinet Connect in order to contact the contributor
Submitted on : Monday, October 25, 2021 - 5:32:58 PM
Last modification on : Friday, December 3, 2021 - 11:42:56 AM

File

grimaldi-21-cbmi.pdf
Files produced by the author(s)

Identifiers

Citation

Antoine Grimaldi, Victor Boutin, Sio-Hoi Ieng, Laurent Perrinet, Ryad Benosman. A homeostatic gain control mechanism to improve event-driven object recognition. 2021 International Conference on Content-Based Multimedia Indexing (CBMI), Jun 2021, Lille, France. pp.1-6, ⟨10.1109/CBMI50038.2021.9461901⟩. ⟨hal-03336554v2⟩

Share

Metrics

Record views

64

Files downloads

105