A Resurgence in Neuromorphic Architectures Enabling Remote Sensing Computation

Vineyard, Craig Michael; Severa, William; Kagie, Matthew J.; Scholand, Andrew J.; Hays, Park

doi:10.1109/spacecomp.2019.00009

Cited by 6 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additional energy is needed to monitor the detectors, which could be accomplished with high speed comparators, which have subns response times and require ∼100 fJ/conversion. Even with this additional energy cost, the performance compares favorably with conventional neural networks where the best systems currently require ∼10 mJ/inference, with projections to ∼100 μJ/inference. , …”

Section: Resultsmentioning

confidence: 99%

“…Even with this additional energy cost, the performance compares favorably with conventional neural networks where the best systems currently require ∼10 mJ/ inference, with projections to ∼100 μJ/inference. 23,24 Robustness to Phase Noise. Fabricating the optical metamaterial requires control over the phases of each aperture.…”

Section: ■ Results and Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Co-Design of Free-Space Metasurface Optical Neuromorphic Classifiers for High Performance

et al. 2021

View full text Add to dashboard Cite

Classification of features in a scene typically requires conversion of the incoming photonic field into the electronic domain. Recently, an alternative approach has emerged whereby passive structured materials can perform classification tasks by directly using freespace propagation and diffraction of light. In this manuscript, we present a theoretical and computational study of such systems and establish the basic features that govern their performance. We show that system architecture, material structure, and input light field are intertwined and need to be co-designed to maximize classification accuracy. Our simulations show that a single layer metasurface can achieve classification accuracy better than conventional linear classifiers, with an order of magnitude fewer diffractive features than previously reported. from multiple layer designs. Finally, we show a trade-off between the number of apertures and fabrication noise.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: ■ Results and Discussionmentioning

confidence: 99%

Co-Design of Free-Space Metasurface Optical Neuromorphic Classifiers for High Performance

et al. 2021

View full text Add to dashboard Cite

show abstract

“…$15.00 DOI: 10.1145/nnnnnnn.nnnnnnn or neural-inspired computer architectures have seen a recent resurgence and are nding new, growing application spaces [18]. While many of the proposed applications focus on size, weight and power (SWaP)-constrained computation [22], a high level of scalability is achievable using neural approaches, and this has lead to increasing interest in large-scale neuromorphic systems to accompany highperformance computing systems [1,8,11,17]. For large-scale or scienti c applications, neuromorphic approaches have been applied to a number of elds and functions including cross-correlation [20], dynamic programming [2], and graph algorithms [10].…”

Section: Introductionmentioning

confidence: 99%

Solving a steady-state PDE using spiking networks and neuromorphic hardware

Darby¹,

Severa²,

Hill³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

e widely parallel, spiking neural networks of neuromorphic processors can enable computationally powerful formulations. While recent interest has focused on primarily machine learning tasks, the space of appropriate applications is wide and continually expanding. Here, we leverage the parallel and event-driven structure to solve a steady state heat equation using a random walk method.e random walk can be executed fully within a spiking neural network using stochastic neuron behavior, and we provide results from both IBM TrueNorth and Intel Loihi implementations. Additionally, we position this algorithm as a potential scalable benchmark for neuromorphic systems.

show abstract