Wednesday, 16 March 2022
Our new paper about the ReckOn chip, developed within the SMALL project, has just been presented at the 2022 International Solid-State Circuits Conference. You can read about our chip and its capabilities here.
Wednesday, 16 March 2022
Monday, 10 January 2022
Our SMALL consortium is wishing you all a prosperous new year!
Sunday, 26 September 2021
Saturday, 25 September 2021
We present a simulation framework of differential-architecture crossbar arrays based on an accurate and comprehensive Phase-Change Memory (PCM) device model, enabling online learning in memristive neuromorphic recurrent neural networks (RNNs). The preprint of our paper is now available here.
Monday, 26 July 2021
Creating biologically realistic models for the underlying computations, especially with spiking neurons and for behaviorally relevant integration time spans, is notoriously difficult. We examine the role of spike frequency adaptation in such computations and find that it has a surprisingly large impact. Details on this research are now published on eLife.
Tuesday, 16 February 2021
PCM-trace is our new neuromorphic building block, which exploits the drift behavior of phase-change materials to implement long lasting eligibility traces, a critical ingredient of spike-based learning rules. The paper describing this novel solution has been accepted to the IEEE International Symposium on Circuits and Systems (ISCAS) 2021 (PDF).
Thursday, 28 January 2021
In real-world applications, learning has to be fast, relying only on a few training examples. How can we achieve such fast learning in spiking neural networks? We have recently published our resarch on this topic in the preprint Revisiting the role of synaptic plasticity and network dynamics for fast learning in spiking neural networks.
Monday, 30 November 2020
Our bio-inspired solution for spiking motor control of a 4 DoF robotic arm has been described in the recently published paper: ED-BioRob: A Neuromorphic Robotic Arm With FPGA-Based Infrastructure for Bio-Inspired Spiking Motor Controllers.
Friday, 30 October 2020
Published a preprint on spatio-temporal learning in deep neural networks.
Thursday, 1 October 2020
Results of our comprehensive study for characterizing the relaxation dynamics of TiOx resistive RAM (RRAM) devices within a predefined volatility framework have been published (Part I: Characterization). In part II of this study, we have also presented a modeling framework that can account for RRAM relaxation characteristics (Part II: Modeling).