DARPA selects Cadence for ML electronic design program

July 26, 2018 //By Ally Winning
DARPA selects Cadence for ML electronic design program
Electronic design automation company Cadence has been chosen by DARPA to use machine learning (ML) techniques to create a unified platform that will lead to a design flow for SoCs, SiPs and PCBs.

The contract is in support the Intelligent Design of Electronic Assets (IDEA) program, one of six new programs within DARPA’s Electronics Resurgence Initiative (ERI). These ERI investments are the next stage to create an automated electronics design capability for the aerospace/defence ecosystem and commercial needs.

The contract lasts four years, and during that time Cadence will be involved in the creation of a machine learning-driven Automatic Generation of Electronic Systems Through Intelligent Collaboration (MAGESTIC) research and development program. The program will introduce greater autonomy to the design process and develop design-intent-driven products. The Cadence-led team includes Carnegie Mellon University and NVIDIA.

The MAGESTIC program is intended to:

- Advance machine learning to develop algorithms to optimize performance

- Extend support for advanced CMOS process nodes

- Automate the routing and tuning of devices to improve reliability, circuit performance, and resilience

- Demonstrate improved power, performance and area (PPA) utilizing machine learning, analytics, and optimization

- Stage the introductions of the technology, allowing the system to learn from the users and teach users how to best achieve desired results.


Related news:
Tortuga tasked by DARPA to bolster chip security
Molecular clock chip can improve smartphone navigation
Autotuning AR glasses with SiN photonics and metasurfaces
Spectre attack prevention added to GrammaTech’s Cyber Hardening Services

Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.