ARM, IBM team on low power analog AI chip
By Nick Flaherty, eeNews Europe (September 9, 2022)
Researchers at ARM and IBM have developed a 14nm analog compute in memory chip for low power always on machine learning.
These always-on perception tasks in IoT applications, dubbed TinyML, require very high energy efficiency. Analog compute-in-memory (CiM) using non-volatile memory (NVM) promises high energy efficiency and self-contained on-chip model storage.
However, analog CiM introduces new practical challenges, including conductance drift, read/write noise, fixed analog-to-digital (ADC) converter gain, etc. These must be addressed to achieve models that can be deployed on analog CiM with acceptable accuracy loss.
Researchers from ARM and IBM Research Zurich looked at the TinyML models for the popular always-on tasks of keyword spotting (KWS) and visual wake words (VWW). The model architectures are specifically designed for analog CiM, and detail a comprehensive training methodology, to retain accuracy in the face of analog issues and low-precision data converters at inference time.
To read the full article, click here
Related Semiconductor IP
- Rad-Hard GPIO, ODIO & LVDS in SkyWater 90nm
- 1.22V/1uA Reference voltage and current source
- 1.2V SLVS Transceiver in UMC 110nm
- 1.8V/3.3V GPIO With I2C Compliant ODIO in GF 55nm
- Verification IP for UALink
Related News
- Nordic Semiconductor and Arm reaffirm partnership with licensing agreement for latest low power processor designs, software platforms, and security IP
- Blumind Harnesses Analog for Ultra Low Power Intelligence
- Blumind reimagines AI processing with breakthrough analog chip
- Blue Cheetah Bunch-of-Wires (BoW) Chiplet Interface Solution Targets Rapid Flexibility, Scalability, and Low Overhead
Latest News
- EXTOLL received GlobalFoundries Award for “Interface IP Partner of the Year”
- AiM Future and Franklin Wireless Sign MOU to Jointly Develop Lightweight AI Model and High-Efficiency 1 TOPS AI SoC Chipset
- GlobalFoundries and Silicon Labs Partner to Scale Industry-Leading Wi-Fi Connectivity
- GlobalFoundries Announces Availability of 22FDX+ RRAM Technology for Wireless Connectivity and AI Applications
- GlobalFoundries Announces Production Release of 130CBIC SiGe Platform for High-Performance Smart Mobile, Communication and Industrial Applications