Fetching the paper…

Reading the bibliography…

2019

Dream Distillation: A Data-Independent Model Compression Framework

Bhardwaj, Kartikeya, Suda, Naveen, Marculescu, Radu

Original
PDF
Post on X

Understand

Model compression is eminently suited for deploying deep learning on IoT-devices.

However, existing model compression techniques rely on access to the original or some alternate dataset.
In this paper, we address the model compression problem when no real data is available, e.g., when data is private.
To this end, we propose Dream Distillation, a data-independent model compression framework.

Dream Distillation: A Data-Independent Model Compression Framework · Around

Built on

Generative adversarial nets
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y · 2014
Earlier work this paper cites.
Distilling the knowledge in a neural network
Original
Hinton, G., Vinyals, O., and Dean, J · 2015
Earlier work this paper cites.
Data-free parameter pruning for deep neural networks
Original
Srinivas, S. and Babu, R. V · 2015
Earlier work this paper cites.
Designing energy-efficient convolutional neural networks using energy-aware pruning
Original
Yang, T.-J., Chen, Y.-H., and Sze, V · 2016
Earlier work this paper cites.

Similar

Knowledge distillation using unlabeled mismatched images
Original
Kulkarni, M., Patil, K., and Karande, S · 2017
Cited alongside, same era.
Data-free knowledge distillation for deep neural networks
Original
Lopes, R., Fenu, S., and Starner, T · 2017
Cited alongside, same era.
Improving the performance of convolutional neural networks via attention transfer
Original
Zagoruyko, S. and Komodakis, N · 2017
Cited alongside, same era.
The building blocks of interpretability
Olah, C · 2018
Cited alongside, same era.

Then

Relaxed quantization for discretized neural networks
Louizos, C. and et al · 2019
Closest in time.
Deepdream
Mordvintsev, A., Olah, C., and Tyka, M · 2019
Closest in time.
Tensorflow lucid
Schubert, L., Olah, C., Mordvintsev, A., Johnson, I., Satyanarayan, A., and et al · 2019
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…