Hesham Mostafa Elsayed

Position:
Postdoc -- ended Jul 2016
Email:
Work phone:
+41766309228
Location:
I obtained my Bachelor's degree in Electrical Engineering from Cairo University in 2007, and my Masters degree in communication electronics from the Technical University of Munich in 2009. I worked for 4 years at Infineon Munich in the technology development group and in the analog and mixed signal verification group before joining INI in 2011.
I am investigating the forms of usable computation that can be obtained from the dynamics of biologically inspired recurrent neural networks. In particular, I am interested in how these networks can solve hard problems like constraint satisfaction and MAP inference problems and how they can use external inputs to update/learn the constraints or the probability distribution defining the problem. I am developing mixed signal (digital/analog), low power VLSI devices comprising biologically realistic spiking neurons and synapses that can be used to directly implement the distributed and parallel forms of computation arising in recurrent networks. I am interested in investigating the potential performance gains that can be obtained by solving the aforementioned hard problems using the low power VLSI devices that I develop as compared to solving them using conventional digital processors.

Publications

2016

2015

2014

  • Mostafa, Hesham and Corradi, Federico and Stefanini, Fabio and Indiveri, Giacomo A Hybrid Analog/Digital Spike-Timing Dependent Plasticity Learning Circuit for Neuromorphic VLSI Multi-Neuron Architectures, ISCAS 854 - 857, 2014
  • Mostafa, Hesham and Indiveri, Giacomo Sequential Activity in Asymmetrically Coupled Winner-Take-All Circuits, Neural Computation, 26:(9) 1973-2004, 2014

2013

© 2017 Institut für Neuroinformatik