Skip to Main Content
White Paper

A hierarchical and tractable mixed-signal verification methodology for first-generation analog AI processors

Mythic AI's methodology for creating their unique analog AI ICs leverages Siemens EDA's Analog FastSPICE and Symphony solutions.

Reading time: 5 minutes
Picture of an integrated circuit chip on a dark bluish purple printed circuit board with traces in every direction lit up in light blue. There are also spots of gold lights in various areas surrounding the chip.
Artificial intelligence (AI) is now the key driving force behind advances in information technology, big data and the internet of things (IoT). It is a technology that is developing at a rapid pace, particularly when it comes to the field of deep learning. Researchers are continually creating new variants of deep learning that expand the capabilities of machine learning. But building systems that are able to use these deep learning models to analyze real-world data presents a major challenge. Silicon cost and energy consumption are major hurdles to teams keen to put deep learning into edge devices, as well as data centers. Conventional digital technology is unable to handle the high compute requirements of AI models that need to run in real-time on cost-effective, low-power hardware. This is a problem that needs to be addressed by a change in technology – a change to a hardware platform that employs analog compute-in-memory (CIM). This is the technology pioneered by Mythic AI, a new generation of AI technology that can propel the next wave of applications that harness deep learning and its many variants.

Share