Seminar & Colloquium
[세미나: 11월 15일(금), 오전 10시] Dr. Sabina Spiga, IMM (Institute for Microelectronics and Microsystems) of the National Research Council (CNR)
Title
Memristive technologies for future bio-inspired computing
Speaker
Dr. Sabina Spiga, Research Director, IMM (Institute for Microelectronics and Microsystems) of the National Research Council (CNR), Italy
* Biography
S. Spiga is currently Research Director at IMM (Institute for Microelectronics and Microsystems) of the National Research Council (CNR), Italy. She is in charge of developing resistive switching non-volatile memories and memristive devices for neuromorphic systems and novel computing paradigms. She is currently member of the IEEE EDS Neuromorphics Technical Committee, of the executive editorial board of the J. Phys D: Applied Physics, and she is associate editor of Frontiers - Neuromorphic Engineering section. S. Spiga has authored more than 150 publications on peer reviewed journals, proceedings and book chapters; and she has been principal investigator for CNR of several European and Italian R&D Projects. Sabina Spiga received the Degree in Physics from the University of Bologna in 1995 and the Ph.D. in Material Science in 2002 from Università di Milano.
| Date | Friday, November 15th, 2024
| Time | 10:00 ~
| Venue | 33동 226호
[Abstract]
Memristive devices (or memristors) have been receiving an increasing interest in a wide range of applications, such as storage class memory, non-volatile logic switch, in-memory computing and neuromorphic computing. In bio-inspired computing systems, memristive devices hold the great promise of novel and versatile dynamic building blocks for hardware neural networks; for instance, memristors can act as dispersed memory elements mimicking synapses in nervous systems, or as stochastic and non-linear elements in neuronal units. Among the proposed technologies, resistance switching memories (RRAMs), based on redox reactions and electrochemical phenomena in oxides and/or solid electrolytes, are very promising because of low power consumption, fast switching times, scalability down to nm scale and CMOS compatibility.
In this talk, I will first introduce the current state of the art and role of memristive systems in neuromorphic computing. In the second part, I will present a brief overview of our results on (i) analogue non-volatile HfOx-based RRAM as synaptic nodes in spiking neural networks, and (ii) Ag/SiOx-based volatile memristors and how to exploit them for neuromorphic functionalities such as short-term plasticity, paired-pulse facilitation and inhibition observed in biological synapses, and integrative function of neurons.
| Host | 장호원 교수(02-880-1720)