Spiking neural networks for real-time inference tasks

09:00-17:00, January 25

Workshop / Overview

In this hands-on tutorial-style workshop, you’ll learn why bringing neural networks back to their biological roots is the way to build ultra-low-power (mW and below) low-latency ML applications for IoT / edge computing.

We’ll give an introduction to spiking neural networks, and how to design spiking NN architectures. We’ll work through two simple sensory applications in audio and vision processing, and tweak and retrain network architectures in order to reduce power consumption.
We’ll also demonstrate deploying spiking networks on ultra-low-power inference hardware from aiCTX.

Workshop / Outcome

At the end of the workshop, you’ll have learned several spiking neural network design approaches, with hands-on experience building and training these architectures using open-source tools.

You’ll get an introduction to using two open-source packages for spiking neural network design, simulation and training, that you can use to build your own applications.
You’ll take home several worked-through Jupyter notebooks to play with further.

Workshop / Difficulty

Intermediate level

Workshop / Prerequisites

  • Intermediate ML knowledge
  • Python experience
  • Own laptop with python and packages preinstalled

Track / Co-organizers

Felix Bauer

R&D Engineer, aiCTX

Dylan Muir

Senior R&D Engineer, aiCTX

Sadique Sheik

Senior Research and Development Engineer, aiCTX

AMLD EPFL 2020 / Workshops

A Conceptual Introduction to Reinforcement Learning

With Bram Vandendriessche, Kevin Smeyers & Katrien Van Meulder

09:00-12:30 January 25

Applied Machine Learning with R

With Dirk Wulff

09:00-17:00 January 25

Augmenting the Web browsing experience using machine learning

With Tudor Avram, Oleksandr Paraska, Vasily Kuznetsov & Levan Tsinadze

09:00-12:30 January 25

AMLD / Global partners