Embedded Systems

Implementation of an Abstract Machine Learning Accelerator Generator based on ACADL

Bachelor’s Thesis / Student Research Project

Abstract

Abstract modeling of HW/SW systems is a relatively new research topic. This technique aims to capture only the essential parameters of software and hardware that influence their timing behavior.

This student project’s goal is to implement a Python-based generator that creates Machine Learning Accelerator descriptions based on the Abstract Computer Architecture Description Language (ACADL). Furthermore, those architectures will be evaluated using different methods for runtime estimation.

An Example of a simple machine learning accelerator modelled with ACADL is presented here:

2x2 Systolic Array in ACADL

Requirements

  • Python
  • Successfully atteded the lecture “Grundlagen der Rechnerarchitektur” and/or “Parallele Rechnerarchitekturen” (optional)
  • Linux (optional)

Contact

Lübeck, Konstantin

Jung, Alexander

Bringmann, Oliver