Scalable Analog Neural-networks (ScAN)

The Defense Advanced Research Projects Agency (DARPA) is soliciting innovative proposals in the following technical areas: the research and development of scalable, robust, and power-efficient analog neural network (NN) architectures and circuits that could directly interface with the analog outputs of conventional sensors. Proposed research should investigate innovative approaches that enable revolutionary advances in science, devices, or systems. Specifically excluded is research that primarily results in evolutionary improvements to the existing state of practice including in-memory-compute (or compute-in-memory) device technology, non-volatile memory development, analog-digital conversion technology, digital accelerator technology, and general-purpose computing technology.

Program Description: The Scalable Analog Neural-networks (ScAN) program will develop new analog NNs that could interface directly with the analog outputs of conventional sensors and demonstrate a three-orders-of-magnitude power reduction over existing solutions. ScAN systems will demonstrate inferencing capabilities of analog NNs while eliminating the need for analog-to-digital converters at the raw sensor level. More information HERE.

PROJECT TIMELINE

Program Name: DARPA Scalable Analog Neural-networks (ScAN)

Solicitation Number: HR001124S0022

Proposal Abstract Due: Date: July 8, 2024 at 4pm

Proposal Due Date: August 22, 2024 at 4pm

Sponsor: DARPA

Anticipated Budget: TBD