NEUROMORPHIC COMPUTING

Success of artificial intelligence (AI) methods typically require large computational times and energy for computation. Recently, computing on neuromorphic chips has shown gains up to 100000× in terms of energy-delay product for certain applications compared to the traditional chips. Despite these promising initial results, extending these successes to a wider range of applications is a major challenge due to limitations of learning algorithms applicable on neuromorphic computing platforms.  Here we focus on this challenge. Our main tool is spiking neural networks (SNNs). SNNs process data using spikes similar to how our brains process information. Using SNNs, we  develop machine learning solutions compatible with neuromorphic hardware. The overall goal is to contribute to data and energy efficient AI using these brain-inspired solutions.

Publications:

S. Karilanova, S. Dey, A. Özçelikkale, State-Space Model Inspired Multiple-Input Multiple-Output Spiking Neurons, Neuro Inspired Computational Elements (NICE) Conference, 2025

S. Karilanova, M.Fabre, E. Neftci, A. Özçelikkale, Zero-Shot Temporal Resolution Domain Adaptation for Spiking Neural Networks,  arXiv:2411.04760, Preprint, 2024

L.Chen,S.Karilanova, S.Chaki, C.Wen, L.Wang, B.Winblad, S.Zhang, A. Özçelikkale, Z.Zhang “Spike-Timing Based Coding in Neuromimetic Tactile System Enables Dynamic Object Classification”. Science. 384(6696):660-665, 2024

Yik et. al, “The neurobench framework for benchmarking neuromorphic computing algorithms and systems”, Nature Communications, 16, vol. 1545, 2025. Arxiv version is here. The related python packages are available at: Neurobench Github Page.