Neuroflow: SW/HW-Interfaces for Neuromorphic Computing

Background

Machine  learning,  big-data  and  internet-of-things are  some  of  the  emerging  computing  applications  which  are extremely  demanding  in  terms  of  storage,  energy  and  performance. While conventional von Neumann architectures are facing significant challenges to cope with such demands, computing-in-memory  architectures  in  general  and  neuro-inspired  architectures, in  particular, represent a  promising  solution  to  overcome these  limitations. This Master Thesis aims to tackle the Software-Hardware interface of novel neuro-inspired accelerators by investigating the mapping and scheduling of state-of-the-art workloads onto memristor-based systolic areas.

Description

In the context of this work, we will invent a novel way of computation in neuromorphic devices. These problems are closely related to existing soltions for Google's TPU-style way of data processing. Their approaches must be translated into the neuromorphic world.

Tasks

Amoung others, these step must be done:

  • Characterize neuromophic processing devices
  • Implement neuromorphic extension for Tensorflow
  • Formulate optimization problems for mapping of compute
  • Propose a neuromorphic mapping and scheduling tool chain
  • Evaluation and verification by means of simulation

Supervisor

Felix Staudigl

Requirements

You will need to have...

  • ... an interest in modelling
  • ... good knowledge of C++ and Python.

In case of interest, please send me an email including the following:

  • Very short motivational letter (4-6 sentences)
  • Latest transcript of records
  • A brief description of your background and CV (if available)