Deep Model Compression and Inference Speedup of Sum-Product Networks on Tensor Trains

Deep Model Compression and Inference Speedup of Sum-Product Networks on Tensor Trains

Ko, Ching Yun (The University of Hong Kong)
Chen, Cong (The University of Hong Kong)
He, Zhuolun (The University of Hong Kong)
Zhang, Yuke (University of South Carolina)
Batselier, K. (TU Delft Team Jan-Willem van Wingerden)
Wong, Ngai (The University of Hong Kong)

2020

Sum-product networks (SPNs) constitute an emerging class of neural networks with clear probabilistic semantics and superior inference speed over other graphical models. This brief reveals an important connection between SPNs and tensor trains (TTs), leading to a new canonical form which we call tensor SPNs (tSPNs). Specifically, we demonstrate the intimate relationship between a valid SPN and a TT. For the first time, through mapping an SPN onto a tSPN and employing specially customized optimization techniques, we demonstrate improvements up to a factor of 100 on both model compression and inference speedup for various data sets with negligible loss in accuracy.

Model compression
sum-product network (SP)
tensor train (TT)

http://resolver.tudelft.nl/uuid:542077fa-62f1-4fb5-8a05-cb381160dc91

https://doi.org/10.1109/TNNLS.2019.2928379

2020-02-09

2162-237X

IEEE Transactions on Neural Networks and Learning Systems, 31 (7), 2665-2671

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Institutional Repository

journal article

© 2020 Ching Yun Ko, Cong Chen, Zhuolun He, Yuke Zhang, K. Batselier, Ngai Wong