M. A. Petrovici, S. Schmitt, J. Klähn, D. Stöckel, A. Schroeder,
G. Bellec, J. Bill, O. Breitwieser, I. Bytschok, A. Grübl, M. Güttler,
A. Hartel, S. Hartmann, D. Husmann, K. Husmann, , S. Jeltsch, V. Karasenko,
M. Kleider, C. Koke, A. Kononov, C. Mauch, P. Müller, J. Partzsch,
T. Pfeil, S. Schiefer, S. Scholze, A. Subramoney, V. Thanasoulis,
B. Vogginger, R. Legenstein, W. Maass, R. Schüffny, C. Mayr, J. Schemmel,
and K. Meier
Despite being originally inspired by the central nervous system, artificial
neural networks have diverged from their biological archetypes as they have
been remodeled to fit particular tasks. In this paper, we review several
possibilites to reverse map these architectures to biologically more
realistic spiking networks with the aim of emulating them on fast, lowpower
neuromorphic hardware. Since many of these devices employ analog components,
which cannot be perfectly controlled, finding ways to compensate for the
resulting effects represents a key challenge. Here, we discuss three
different strategies to address this problem: the addition of auxiliary
network components for stabilizing activity, the utilization of inherently
robust architectures and a training method for hardwareemulated networks that
functions without perfect knowledge of the system’s dynamics and
parameters. For all three scenarios, we corroborate our theoretical
considerations with experimental results on accelerated analog neuromorphic
platforms.