Reintroducing Straight-Through Estimators as Principled Methods for Stochastic Binary Networks
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F21%3A00357178" target="_blank" >RIV/68407700:21230/21:00357178 - isvavai.cz</a>
Result on the web
<a href="https://doi.org/10.1007/978-3-030-92659-5_7" target="_blank" >https://doi.org/10.1007/978-3-030-92659-5_7</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1007/978-3-030-92659-5_7" target="_blank" >10.1007/978-3-030-92659-5_7</a>
Alternative languages
Result language
angličtina
Original language name
Reintroducing Straight-Through Estimators as Principled Methods for Stochastic Binary Networks
Original language description
Training neural networks with binary weights and activa-tions is a challenging problem due to the lack of gradients and difficulty ofoptimization over discrete weights. Many successful experimental resultshave been achieved with empirical straight-through (ST) approaches,proposing a variety of ad-hoc rules for propagating gradients throughnon-differentiable activations and updating discrete weights. At the sametime, ST methods can be truly derived as estimators in the stochasticbinary network (SBN) model with Bernoulli weights. We advance thesederivations to a more complete and systematic study. We analyze proper-ties, estimation accuracy, obtain different forms of correct ST estimatorsfor activations and weights, explain existing empirical approaches andtheir shortcomings, explain how latent weights arise from the mirrordescent method when optimizing over probabilities. This allows to rein-troduce ST methods, long known empirically, as sound approximations,apply them with clarity and develop further improvements.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
<a href="/en/project/EF16_019%2F0000765" target="_blank" >EF16_019/0000765: Research Center for Informatics</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2021
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Proceedings of the 43rd DAGM German Conference (DAGM GCPR 2021)
ISBN
978-3-030-92658-8
ISSN
0302-9743
e-ISSN
1611-3349
Number of pages
16
Pages from-to
111-126
Publisher name
Springer
Place of publication
Cham
Event location
Bonn
Event date
Sep 28, 2021
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—