We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.AR

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Hardware Architecture

Title: A Custom 7nm CMOS Standard Cell Library for Implementing TNN-based Neuromorphic Processors

Abstract: A set of highly-optimized custom macro extensions is developed for a 7nm CMOS cell library for implementing Temporal Neural Networks (TNNs) that can mimic brain-like sensory processing with extreme energy efficiency. A TNN prototype (13,750 neurons and 315,000 synapses) for MNIST requires only 1.56mm2 die area and consumes only 1.69mW.
Comments: This work is dated and will be superseded by a forthcoming work
Subjects: Hardware Architecture (cs.AR); Emerging Technologies (cs.ET); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
Cite as: arXiv:2012.05419 [cs.AR]
  (or arXiv:2012.05419v2 [cs.AR] for this version)

Submission history

From: Harideep Nair [view email]
[v1] Thu, 10 Dec 2020 02:31:57 GMT (2483kb)
[v2] Fri, 4 Jun 2021 22:01:18 GMT (0kb,I)

Link back to: arXiv, form interface, contact.