Topological acoustic synapse for high-dimensional neuromorphic computing
Open Access
- 12 June 2026
- journal article
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (24) , eaec6633
Abstract
The human brain performs complex, high-dimensional (HD) computations, such as causal reasoning, counterfactual thinking, and abstraction, with ~10 11 neurons while consuming ~20 watts of power. Neuromorphic computing seeks similar efficiency, but current devices face bottlenecks in bandwidth, energy, wiring, footprint, and reliability that limit scalability. Here, we introduce the topological acoustic synapse (TAS), an acoustic-wave neuromorphic device that circumvents these limits by mapping information in multivariate state spaces. A single TAS generates and manipulates numerous computing channels that operate independently and in parallel. The TAS leverages nonlinear interactions to emulate biorealistic neuromorphic functionalities, including reconfigurable synaptic plasticity, neuromodulation, and hybrid analog-digital control. In classification tasks, a TAS handles multiple inputs simultaneously and generates various outputs, converging 20% faster while using 60% fewer parameters and at least an order of magnitude less power than state-of-the-art electrical devices. This work establishes the first acoustic synapse with parallel HD computing capabilities, presenting a scalable paradigm for neuromorphic hardware with high computational density.Keywords
Funding Information
- National Science Foundation (2242925)
- National Science Foundation (2441746)
- Air Force Office of Scientific Research (FA9550-26-1-0326)
This publication has 66 references indexed in Scilit:
- Coupled Ferroelectric‐Photonic Memory in a Retinomorphic Hardware for In‐Sensor ComputingAdvanced Science, 2024
- Practical implementation of a scalable discrete Fourier transform using logical phi-bits: nonlinear acoustic qubit analoguesQuantum Studies: Mathematics and Foundations, 2023
- Integrated Memory Devices Based on 2D MaterialsAdvanced Materials, 2022
- Recent Progress on Emerging Transistor‐Based Neuromorphic DevicesAdvanced Intelligent Systems, 2021
- Breaking the von Neumann bottleneck: architecture-level processing-in-memory technologyScience China Information Sciences, 2021
- Mutually Reinforced Polymer–Graphene Bilayer Membranes for Energy‐Efficient Acoustic TransductionAdvanced Materials, 2020
- Sound Topology, Duality, Coherence and Wave-Mixing2017
- A material history of bitsJournal of the American Society for Information Science and Technology, 2011
- Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled‐up primate brainJournal of Comparative Neurology, 2009
- Hyperdimensional Computing: An Introduction to Computing in Distributed Representation with High-Dimensional Random VectorsCognitive Computation, 2009