Explore projects

https://iffmd.fz-juelich.de/RhXA4J_JTwebjYKG35TbuQ#

Electron transport calculation code based on the real-space finite-difference formalism within the framework of the density functional theory

Convert Fortran modules to C-headers. Taken over from https://github.com/sharifmarat/fortran_to_c_headers

Title: An introduction to training algorithms for neuromorphic computing and on-line learning Abstract: Training neural networks implemented in neuromorphic hardware is challenging due to the dynamic, sparse, and local nature of the computations. This tutorial will describe some established gradient-based solutions to address these challenges in the context of real-valued recurrent neural networks and spiking neural networks. Insights into gradient-based training algorithms and associated autodifferentiation methods lead to online synaptic plasticity rules and the necessary assumptions to implement them in in-memory computing devices. The tutorial will conclude with methods that can be used to improve and optimize learning algorithms using meta-learning and other meta-optimization approaches.