Michael Pfeiffer

Position:
Group Leader -- ended Oct 2016
Email:
Work phone:
+41 44 635 30 45
Home page:
I studied mathematics and computer science at TU Graz in Austria, where I also obtained my PhD. Initially working on machine learning and artificial intelligence, I became more and more fascinated by the "real" intelligence in our brains, and how biology still outperforms modern computing technology in most real-world scenarios. Currently my work focuses on algorithmic approaches for understanding computation and learning in the brain, and linking observations from biology to mathematical formalisms such as probabilistic inference. I evaluate these theories and event-based algorithms in neuromorphic engineering applications, in particular for event-based vision. My goal is to advance the current state-of-the-art in artificial systems in terms of efficiency and performance, by applying the principles we learn from neuroscience. At INI, I am also coordinating the "Neural Systems and Computation" MSc program.

Publications

2017

  • Neil D., Pfeiffer M., and Liu S-C Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences, Proc. of NIPS 2016, 2017

2016

2015

2014

2013

2012

  • Lee, J. and Delbruck, T. and Park, P.K.J. and Pfeiffer, M. and Shin, C.W. and Ryu, H. and Kang, B.C. Live demonstration: Gesture-Based remote control using stereo pair of dynamic vision sensors, IEEE Intl. Symp. on Circuits and Systems (ISCAS 2012) 736-740, 2012

2011

© 2017 Institut für Neuroinformatik