NeuroPDE: A Neuromorphic PDE Solver Based on Spintronic and Ferroelectric Devices

2025-07-07

AI Summary

Traditional von Neumann architectures lack intrinsic hardware-level randomness, limiting the efficiency of random-walk-based partial differential equation (PDE) solvers. To address this, we propose NeuroPDE—the first brain-inspired PDE solver integrating spintronic neurons with ferroelectric synapses. It employs probabilistic spiking neurons driven by the spin Hall effect and analog, non-volatile synaptic weights implemented via ferroelectric-gate transistors, directly harnessing device-level stochasticity for hardware-native probabilistic computation. This work achieves the first synergistic integration of spintronic and ferroelectric devices for PDE solving. Experimental results demonstrate that NeuroPDE solves diffusion equations with an error variance below 1×10⁻². Compared to state-of-the-art CMOS neuromorphic chips, it achieves 3.48–315× speedup and 2.7–29.8× energy-efficiency improvement.

Technology Category

• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Security and Privacy: Privacy-enhancing technologies

Abstract

In recent years, new methods for solving partial differential equations (PDEs) such as Monte Carlo random walk methods have gained considerable attention. However, due to the lack of hardware-intrinsic randomness in the conventional von Neumann architecture, the performance of PDE solvers is limited. In this paper, we introduce NeuroPDE, a hardware design for neuromorphic PDE solvers that utilizes emerging spintronic and ferroelectric devices. NeuroPDE incorporates spin neurons that are capable of probabilistic transmission to emulate random walks, along with ferroelectric synapses that store continuous weights non-volatilely. The proposed NeuroPDE achieves a variance of less than 1e-2 compared to analytical solutions when solving diffusion equations, demonstrating a performance advantage of 3.48x to 315x speedup in execution time and an energy consumption advantage of 2.7x to 29.8x over advanced CMOS-based neuromorphic chips. By leveraging the inherent physical stochasticity of emerging devices, this study paves the way for future probabilistic neuromorphic computing systems.

Problem

• Hardware-intrinsic randomness lacking in conventional PDE solvers
• Neuromorphic PDE solver design using spintronic and ferroelectric devices
• Improving performance and energy efficiency in solving diffusion equations

Innovation

• Uses spintronic devices for probabilistic transmission
• Employs ferroelectric synapses for non-volatile storage
• Leverages physical stochasticity for neuromorphic computing