Controlling ReRAM’s Switching Characteristics with Shadow Memory for Continual Learning
We propose integrating a "shadow" memristor with a synaptic memristor to adjust switching characteristics of the latter, e.g., by scaling programming voltages using a voltage divider configuration. Such adjustment allows tightening device-to-device variations in switching behavior across a chip, an essential requirement for implementing memristor-based fast (few-shot, real-time) learning neuromorphic systems. We experimentally validated our approach on two exemplary fast learning scenarios to show that it significantly improves their performance. Additionally, we demonstrate that shadow memristors help avoid catastrophic forgetting by shifting switching threshold voltages of sensitive synaptic memristors. Therefore, the proposed approach enables two key features of continual learning.
Authors
- Thomas Van Vaerenbergh (ORCID: https://orcid.org/0000-0002-7301-8610)
- Sai Sukruth Bezugam (ORCID: https://orcid.org/0000-0003-0750-6568)
- Dmitri B. Strukov
- Jim Ignowski (ORCID: https://orcid.org/0000-0001-5091-3674)
- Robert Legenstein (ORCID: https://orcid.org/0000-0002-8724-5507)
- George Hutchinson (ORCID: https://orcid.org/0009-0000-0413-4362)
- Tanmoy Bhattacharya (ORCID: https://orcid.org/0000-0002-6453-8972)
- Raymond G. Beausoleil (ORCID: https://orcid.org/0000-0003-1139-1660)
- Xia Sheng (ORCID: https://orcid.org/0000-0002-8591-0841)
- Giacomo Pedretti (ORCID: https://orcid.org/0000-0002-4501-8672)
- Horst Petschenig (ORCID: https://orcid.org/0000-0003-3106-7008)
- Sanghyeon Choi (ORCID: https://orcid.org/0000-0002-8356-5749)
Institutions
- Hewlett-Packard (United States) (US)
- University of California, Santa Barbara (US)
- Graz University of Technology (AT)
Publication Details
- Published
- 2025-05-18
- DOI
- https://doi.org/10.1109/imw61990.2025.11026884
- Citations
- 1
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.73