TY - CONF
AU - Lohoff, Jamie
AU - Neftci, Emre
TI - Finding new bio-plausible Learning Rules using Deep Reinforcement Learning
PB - RWTH Aachen
M1 - FZJ-2024-01020
PY - 2023
AB - Gradient-based learning is still the best bet when training spiking neural networks on supervised tasks. Although backpropagation, the state-of-the-art in modern AI, is not bio-plausible, there exists a wide range of approximations with this property that achieve competitive performance, e.g. e-prop[1,2]. We propose a new framework called AlphaGrad that could find more such learning rules by systematically exploring the search space using Deep Reinforcement Learning and methods from Automatic Differentiation(AD).
T2 - nternational conference on neuromorphic, natural and physical computing
CY - 25 Oct 2023 - 27 Oct 2023, Hanover (Germany)
Y2 - 25 Oct 2023 - 27 Oct 2023
M2 - Hanover, Germany
LB - PUB:(DE-HGF)24
DO - DOI:10.34734/FZJ-2024-01020
UR - https://juser.fz-juelich.de/record/1021872
ER -