Computer Architectures for AI Applications
In this course the focus will be on the specifics of hardware design and architectures for AI applications. After the overview of the standard design techniques and common computing architectures, the additional requirements of AI will be discussed. Based on this, the specific architectures and design methods increasing the efficiency of the computation will be discussed. Finally, this course will include also an introduction to the emerging and novel architectures and technologies that could have significant impact in the future.
Here is the detailed list of topics:
- Introduction in VLSI design and computer architectures
- State of the art processor architecture, Example RISC-V
- Limitations of classical architectures for AI applications
- Accelerators architectures: GPUs, MAC arrays
- Neuromorphic Architectures (TrueNorth, Loihi, Spinnaker), asynchronous design
- Emerging architectures: In-Memory-Computing (example RRAM)
Lecturer | Prof. Miloš Krstić Anselm Breitenreiter |
Type of Lecture | Lecture (2 SWS) Labs (2 SWS) |
Target Group | 4th Semester |
Affiliation | Technical Computer Science |
Prerequisites | Technische Grundlagen der Informatik or Informationsverarbeitung |
Language | German (Lecture) English (Slides) |
Examination | Oral Exam |
Points | 6 LP |
To register for this lecture, please use Puls