Neuromorphic computing with energy efficient quantum materials

ORAL

Abstract

Neuromorphic computing has been proposed as a solution for society's exponentially growing need for energy efficient computation. In the era of artificial intelligence, neuromorphic computing aims to provide a hardware analog for a human brain's capacity for memory and computation. One possible energy-efficient platform for this is the quantum phenomena in perovskite nickelates to simulate neurons, synapses and an overall interconnected network with reconfigurable synaptic plasticity. The technology for such aspirations relies on two former accomplishments. First, hydrogenation of rare earth nickelates, which has been linked to significant magnification of the material's local resistivity. Second, focused helium ion modification, which was shown to effectively suppress local metal-insulator transitions (MIT). Intentional coupling of these two features opens up the opportunity to design a synapse-like unit on an epitaxially grown nickelate thin film, with synthetically implemented excitatory and inhibitory behaviors required for neuron-synapse interactions. Therefore, a complex network of such units is hypothesized to function as a trainable, self sustained neural network, with some units inhibiting, while others facilitating the flow of information.

*DOE EFRC (Department of Energy - Energy Frontier Research Center)

–

Publication: Abbott, L. F., & Nelson, S. B. (2000). Synaptic plasticity: Taming the beast. Nature Neuroscience, 3(11s), 1178–1183. https://doi.org/10.1038/81453

Bisht, R. S., Park, J., Yu, H., Wu, C., Tilak, N., Rangan, S., Park, T. J., Yuan, Y., Das, S., Goteti, U., Yi, H. T., Hijazi, H., Al-Mahboob, A., Sadowski, J. T., Zhou, H., Oh, S., Andrei, E. Y., Allen, M. T., Kuzum, D., … Ramanathan, S. (2023). Spatial Interactions in Hydrogenated Perovskite Nickelate Synaptic Networks. Nano Letters, 23(15), 7166–7173. https://doi.org/10.1021/acs.nanolett.3c02076

Goteti, U. S., Zaluzhnyy, I. A., Ramanathan, S., Dynes, R. C., & Frano, A. (2021). Low-temperature emergent neuromorphic networks with correlated oxide devices. Proceedings of the National Academy of Sciences of the United States of America, 118(35). https://doi.org/10.1073/pnas.2103934118

Luibrand, T., Bercher, A., Rocco, R., Tahouni-Bonab, F., Varbaro, L., Rischau, C. W., Domínguez, C., Zhou, Y., Luo, W., Bag, S., Fratino, L., Kleiner, R., Gariglio, S., Koelle, D., Triscone, J. M., Rozenberg, M. J., Kuzmenko, A. B., Guénon, S., & del Valle, J. (2023). Characteristic length scales of the electrically induced insulator-to-metal transition. Physical Review Research, 5(1). https://doi.org/10.1103/PhysRevResearch.5.013108

Wu, J., D'hollande, A., Du, H., & Rozenberg, M. (n.d.). A dynamic neuro-synaptic hardware platform for Spiking Neural Networks.

Yuan, Y., Patel, R. K., Banik, S., Reta, T. B., Bisht, R. S., Fong, D. D., Sankaranarayanan, S. K. R. S., & Ramanathan, S. (2024). Proton Conducting Neuromorphic Materials and Devices. In Chemical Reviews. American Chemical Society. https://doi.org/10.1021/acs.chemrev.4c00071

Zhang, H. T., Park, T. J., Zaluzhnyy, I. A., Wang, Q., Wadekar, S. N., Manna, S., Andrawis, R., Sprau, P. O., Sun, Y., Zhang, Z., Huang, C., Zhou, H., Zhang, Z., Narayanan, B., Srinivasan, G., Hua, N., Nazaretski, E., Huang, X., Yan, H., … Ramanathan, S. (2020). Perovskite neural trees. Nature Communications, 11(1). https://doi.org/10.1038/s41467-020-16105-y

Presenters

  • Nika Bondar

    • California Polytechnic State University

Authors

  • Nika Bondar

    • California Polytechnic State University

  • Robin Glefke

    • University of California, San Diego

  • Rourav Basak

    • University of California, San Diego

  • Ivan K Schuller

    • University of California, San Diego

  • Alex Frano

    • University of California, San Diego

  • Shriram Ramanathan, PhD

    • Rutgers University