AI-Pro: Responsible AI-Processor Design
Sustainability |
Research Goals
Visionary Research
The Chair of AI Processor Design (AI-Pro) plays a key role in shaping the future of AI processors by exploring novel computer architectures that go beyond traditional computing and the von Neumann paradigm. The research aims to integrate AI chips into every aspect of our daily lives, thereby revolutionizing the way we interact with technology and transforming the structure of our societies.
AI-Pro strives for innovation and leverages the potential of emerging technologies. It follows a synergistic approach that integrates beyond-CMOS technologies (CMOS: complementary metal–oxide–semiconductor) with brain-inspired computing. In 2023, the team succeeded in developing a high-performance in-memory computing approach that can significantly accelerate deep learning applications. This novel approach introduces a new computational paradigm based on ferroelectric field-effect transistors (FeFETs). FeFETs are a type of non-volatile memory that can improve the efficiency and speed of data processing.
Advancing Sustainability
With this work, AI-Pro aims to drastically reduce the CO2 footprint of AI and promote a more sustainable and environmentally friendly future. Current trends in AI are driving a massive increase in energy demand and global CO2 emissions. Improving efficiency therefore leads to a reduction in the CO2 footprint.
Researchers
Lead Researcher: Prof. Dr.-Ing. Hussam Amrouch at the Professorship of AI Processor Design