Search NASA⌕ Search
Home/Archive/DOE OSTI

DOE OSTI · 2588895

Reconfigurable neuromorphic components and algorithms for next-generation artificial intelligence

Zutter, Brian Travis [Sandia National Lab. (SNL-CA), Livermore, CA (United States)]·Oh, Sangheon [Sandia National Lab. (SNL-CA), Livermore, CA (United States)] (ORCID:0000000323717410)·Brown, Timothy David [Sandia National Lab. (SNL-CA), Livermore, CA (United States)]·Anderson, Jillian [Stanford Univ., CA (United States)]·Beltran, Saul Perez [Texas A & M Univ., College Station, TX (United States)]·Arenas Blanco, Brayan A. [Texas A & M Univ., College Station, TX (United States)]·Lopez-Meza, Andres [Texas A & M Univ., College Station, TX (United States)]·Christensen, Adam S. [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)]·Bishop, Sean Robert [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)] (ORCID:0000000213331859)·Finnegan, Patrick Sean [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)]·Ievlev, Anton V. [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)]·Woo, Kyung Seok [Sandia National Lab. (SNL-CA), Livermore, CA (United States)]·Sugar, Joshua Daniel [Sandia National Lab. (SNL-CA), Livermore, CA (United States)] (ORCID:0009000245752807)·Li, Yiyang [Univ. of Michigan, Ann Arbor, MI (United States)]·Fuller, Elliot James [Sandia National Lab. (SNL-CA), Livermore, CA (United States)]·Balbuena, Perla B. [Texas A & M Univ., College Station, TX (United States)]·Williams, R. Stanley [Texas A & M Univ., College Station, TX (United States)]·Kumar, Suhas [Sandia National Lab. (SNL-CA), Livermore, CA (United States)]·Talin, Albert Alec [Sandia National Lab. (SNL-CA), Livermore, CA (United States)] (ORCID:000000021102680X)
View original source ↗Read source document ↗Export metadata ↓

Abstract

Digital transistor-based general-purpose hardware (e.g., central processing units) is the dominant solution to support both traditional computing (logic, arithmetic, etc.) as well as modern artificial intelligence. State-of-the-art research has shown feasibility of post-digital physics-based neuromorphic hardware, which is hypothesized to support artificial intelligence algorithms with orders-of-magnitude improved time/energy efficiencies. But such research has not been widely deployed mainly because of such novel hardware’s extreme application-specificity, and the dominance of low-cost general-purpose (but inefficient) digital hardware. To make use of the novel algorithms and the superlative performance of physics-based hardware, we need to identify scientific principles that can enable generality in physics-based hardware. This work resulted in two important broad outcomes – first, we demonstrate fully reconfigurable neuromorphic components, and second, we demonstrate a viable artificial intelligence learning algorithm that can exploit the functioning of neuromorphic hardware. We demonstrate up to five orders of magnitude improvement in energy efficiency compared to the best general-purpose digital hardware.

Explore related subjects

97 MATHEMATICS AND COMPUTING

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Zutter, Brian Travis [Sandia National Lab. (SNL-CA), Livermore, CA (United States)], Oh, Sangheon [Sandia National Lab. (SNL-CA), Livermore, CA (United States)] (ORCID:0000000323717410), Brown, Timothy David [Sandia National Lab. (SNL-CA), Livermore, CA (United States)], Anderson, Jillian [Stanford Univ., CA (United States)], Beltran, Saul Perez [Texas A & M Univ., College Station, TX (United States)], Arenas Blanco, Brayan A. [Texas A & M Univ., College Station, TX (United States)], Lopez-Meza, Andres [Texas A & M Univ., College Station, TX (United States)], Christensen, Adam S. [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)], Bishop, Sean Robert [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)] (ORCID:0000000213331859), Finnegan, Patrick Sean [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)], Ievlev, Anton V. [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)], Woo, Kyung Seok [Sandia National Lab. (SNL-CA), Livermore, CA (United States)], Sugar, Joshua Daniel [Sandia National Lab. (SNL-CA), Livermore, CA (United States)] (ORCID:0009000245752807), Li, Yiyang [Univ. of Michigan, Ann Arbor, MI (United States)], Fuller, Elliot James [Sandia National Lab. (SNL-CA), Livermore, CA (United States)], Balbuena, Perla B. [Texas A & M Univ., College Station, TX (United States)], Williams, R. Stanley [Texas A & M Univ., College Station, TX (United States)], Kumar, Suhas [Sandia National Lab. (SNL-CA), Livermore, CA (United States)], Talin, Albert Alec [Sandia National Lab. (SNL-CA), Livermore, CA (United States)] (ORCID:000000021102680X). 2025-11-01. Reconfigurable neuromorphic components and algorithms for next-generation artificial intelligence. https://doi.org/10.2172/2588895

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

DOE OSTI2026

TANTE: Time-adaptive operator learning via neural Taylor expansion

Operator learning for time-dependent partial differential equations (PDEs) has seen rapid progress in recent years, enabling efficient approximation of complex spatiotemporal dynamics. However, most existing methods rely on fixed time step sizes during rollout, which limits their ability to adapt to varying temporal complexity and often leads to error accumulation. In this work, we propose the Time-Adaptive Transformer with Neural Taylor Expansion (TANTE), a novel operator-learning framework that produces continuous-time predictions with adaptive step sizes. TANTE predicts future states by performing a Taylor expansion at the current state, where neural networks learn both the higher-order temporal derivatives and the local radius of convergence. This allows the model to dynamically adjust its rollout based on the local behavior of the solution, thereby reducing cumulative error and improving computational efficiency. We demonstrate the effectiveness of TANTE across a wide range of PDE benchmarks, achieving superior accuracy and adaptability compared to fixed-step baselines, delivering accuracy gains of 60-80 % and speed-ups of 30-40 % at inference time.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Structured illumination for surface-resolved grazing-incidence X-ray scattering

Grazing-incidence (GI) scattering techniques are widely used to characterize thin films, offering high surface sensitivity and insight into morphology and structure. However, these approaches typically provide statistical averaged information due to elongated footprint or limited spatial resolution due to beam size. Here we introduce a method that combines structured illumination with GI X-ray scattering and leverages our computational imaging approach to resolve local structural details. We demonstrate that our method captures local features of an organic semiconductor thin film without the need for sample rotation as in tomography. The method expands GI techniques from statistical averaging to high-resolution imaging, thereby providing the capability for detailed analysis of local material properties, such as domain shape, orientation and polymorphism, which are critical for advancing material design towards more efficient and tailored materials.

97 MATHEMATICS AND COMPUTING↗