High-Level Synthesis Enables the Next Generation of Edge AI Accelerators

AI is becoming pervasive. But the ever increasing complexity is a challenge for IoT systems. Achieving the highest levels of performance and efficiency in edge AI means going beyond software and off the shelf hardware. Bespoke hardware accelerators in FPGA or ASICs can deliver much higher performance while consuming less energy. Building these accelerators with High-Level Synthesis slashes design and verification effort, minimizing schedule and execution risks.