Introduction to Neuromorphic Computing and the SpiNNaker Hardware Platform

Event content

Neuromorphic computing is a computing paradigm that combines brain-inspired algorithms and associated hardware architectures. While it has been developed for decades, the field is recently gaining strong traction due to its possible applications in computational neuroscience and AI. Neuromorphic computing is based on the behavior of biological spiking neurons, and is well suited to implement and study the dynamics of neuronal networks, graph and optimization problems, and specific AI/machine learning applications. A key property of neuromorphic hardware is that it often exhibits tremendously lower power consumption when compared to traditional hardware and algorithms, which is a crucial concern in today's power hungry computational workloads.

In this course, you will learn about the basic concepts of neuromorphic computing, as well as the different libraries available to perform neuromorphic simulations. Furthermore, you will be among the first to gain hands-on access to the novel SpiNNaker-2 hardware architecture, and you may get to apply neuromorphic approaches to a series of problems including simulating neuronal networks and machine learning tasks.

This course should be of interest to those working with neuron and network models in neuroscience, AI/machine learning, optimization problems, and to those interested in developing new algorithms and working with novel hardware in general. Only basic knowledge of Python is required, and an idea of how biological neurons operate may be helpful.

Learning goal

• Learn what neuromorphic computing is, and what sort of problems it can solve
• Get to know different hardware systems used in neuromorphic computing
• Gain access to the novel neuromorphic computing platform SpiNNaker-2, and learn how to perform basic neuromorphic simulations with it

Information about the event

Max. participants

50

Requirements

• Basic Python knowledge
• Knowledge of how biological neurons function is an advantage but not necessary

Speakers

Dr. Jannik Luboeinski

Dr. Martin Leandro Paleico