% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@INPROCEEDINGS{Boeshertz:1038063,
author = {Boeshertz, Gauthier and Indiveri, Giacomo and Nair, Manu
and Renner, Alpha},
title = {{A}ccurate {M}apping of {RNN}s on {N}euromorphic {H}ardware
with {A}daptive {S}piking {N}eurons},
publisher = {IEEE},
reportid = {FZJ-2025-01112},
pages = {376-380},
year = {2024},
abstract = {Thanks to their parallel and sparse activity features,
recurrent neural networks (RNNs) are well-suited for
hardware implementation in low-power neuromorphic hardware.
However, mapping rate-based RNNs to hardware-compatible
spiking neural networks (SNNs) remains challenging. Here, we
present a ΣΔ−low-pass RNN (lpRNN): an RNN architecture
employing an adaptive spiking neuron model that encodes
signals using ΣΔ-modulation and enables precise mapping.
The ΣΔ-neuron communicates graded values using spike
timing, and the dynamics of the IpRNN are set to match
typical timescales for processing natural signals, such as
speech. Our approach integrates rate and temporal coding,
offering a robust solution for efficient and accurate
conversion of RNNs to SNNs. We demonstrate the
implementation of the IpRNN on Intel's neuromorphic research
chip Loihi, achieving state-of-the-art classification
results on audio benchmarks using 3-bit weights. These
results call for a deeper investigation of recurrency and
adaptation in event-based systems, which may lead to
insights for edge computing applications where
power-efficient real-time inference is required.},
month = {Jul},
date = {2024-07-30},
organization = {2024 International Conference on
Neuromorphic Systems (ICONS), Arlington
(VA), 30 Jul 2024 - 2 Aug 2024},
cin = {PGI-15},
cid = {I:(DE-Juel1)PGI-15-20210701},
pnm = {5234 - Emerging NC Architectures (POF4-523)},
pid = {G:(DE-HGF)POF4-5234},
typ = {PUB:(DE-HGF)8},
UT = {WOS:001462433900056},
doi = {10.1109/ICONS62911.2024.00064},
url = {https://juser.fz-juelich.de/record/1038063},
}