Abstract
In this work, we demonstrate an implementation of Deep Convolution Generative Adversarial Network (DCGAN), into a Field Programmable Gate Array (FPGA). In order to implement the DCGAN, we modified the DCGAN model with binary weights and activations, and with using integer-valued operations in the forwarding path (train-time and run-time). We call the modified one as Binary-DCGAN (B-DCGAN). Using the B-DCGAN, we do a feasibility study of FPGA’s characteristics and performance for Deep Learning. Because the binarization and using integer-valued operation reduce the memory capacity and the number of the circuit gates, it is very effective for FPGA implementation. On the other hand, these reductions in the model might decrease the quality of the generated data. So we investigate the influence of these reductions.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abdelouahab, K., Pelcat, M., Berry, F., Sérot, J.: Accelerating CNN inference on FPGAs: a survey. Research report, Université Clermont Auvergne; Institut Pascal, Clermont Ferrand; IETR/INSA Rennes, January 2018. https://hal.archives-ouvertes.fr/hal-01695375
Al-Rfou, R., et al.: Theano: a Python framework for fast computation of mathematical expressions. arXiv e-prints abs/1605.02688, May 2016. http://arxiv.org/abs/1605.02688
Cheng, Y., Wang, D., Zhou, P., Zhang, T.: A survey of model compression and acceleration for deep neural networks. arXiv e-prints, October 2017
Courbariaux, M., Hubara, I., Soudry, D., El-Yaniv, R., Bengio, Y.: Binarized neural networks: training deep neural networks with weights and activations constrained to \(+1\) or \(-1\). arXiv e-prints, February 2016
Dieleman, S., et al.: Lasagne: first release, August 2015. https://doi.org/10.5281/zenodo.27878
LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)
Mirza, M., Osindero, S.: Conditional generative adversarial nets. arXiv e-prints, November 2014
Radford, A., Metz, L., Chintala, S.: Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv e-prints, November 2015
Umuroglu, Y., et al.: FINN: a framework for fast, scalable binarized neural network inference. arXiv e-prints, December 2016
Acknowledgements
We would like to thank the UEC Shouno lab (http://daemon.inf.uec.ac.jp/ja/) members: Satoshi Suzuki and Aiga Suzuki for theoretical and technical discussion; Seigo Kawamura, Kurosaka Mamoru, Yoshihiro Kusano, Toya Teramoto and Akihiro Endo for kind technical assistance and humor. We thank Kazuhiko Yoshihara and all the members of Open Stream, Inc. We also thank the developers of Theano, Lasagne, and Python environment. This work is supported under the funds of Open Stream, Inc. (https://www.opst.co.jp/).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Terada, H., Shouno, H. (2019). B-DCGAN: Evaluation of Binarized DCGAN for FPGA. In: Gedeon, T., Wong, K., Lee, M. (eds) Neural Information Processing. ICONIP 2019. Lecture Notes in Computer Science(), vol 11953. Springer, Cham. https://doi.org/10.1007/978-3-030-36708-4_5
Download citation
DOI: https://doi.org/10.1007/978-3-030-36708-4_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-36707-7
Online ISBN: 978-3-030-36708-4
eBook Packages: Computer ScienceComputer Science (R0)