Notcompressing neural networks themselevs, but rather using them to compress other things, in the sense of using them to reconstruct things with low error, especially in concert with existing noisy data transmission pipelines. Of course, we might try to do both at once and then think about minimum description length. that might be interesting.
References
Ananthabhotla, Ishwarya, Sebastian Ewert, and Joseph A. Paradiso. 2019. βTowards a Perceptual Loss: Using a Neural Network Codec Approximation as a Loss for Generative Audio Models.β In Proceedings of the 27th ACM International Conference on Multimedia, 1518β25. Nice France: ACM.
Collobert, Ronan, Awni Hannun, and Gabriel Synnaeve. 2019. βA Fully Differentiable Beam Search Decoder.β In Proceedings of the 36th International Conference on Machine Learning, 1341β50. PMLR.
Eusebio, Paulo, Joao Ascenso, and Fernando Pereira. 2021. βOptimizing an Image Coding Framework with Deep Learning-Based Pre- and Post-Processing.β In 2020 28th European Signal Processing Conference (EUSIPCO), 506β10. Amsterdam, Netherlands: IEEE.
Guleryuz, Onur G., Philip A. Chou, Hugues Hoppe, Danhang Tang, Ruofei Du, Philip Davidson, and Sean Fanello. 2021. βSandwiched Image Compression: Wrapping Neural Networks Around A Standard Codec.β In 2021 IEEE International Conference on Image Processing (ICIP), 3757β61. Anchorage, AK, USA: IEEE.
Klopp, Jan P., Keng-Chi Liu, Liang-Gee Chen, and Shao-Yi Chien. 2021. βHow to Exploit the Transferability of Learned Image Compression to Conventional Codecs.β In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 16160β69. Nashville, TN, USA: IEEE.
Qiu, Kaitian, Lu Yu, and Daowen Li. 2021. βCodec-Simulation Network for Joint Optimization of Video Coding With Pre- and Post-Processing.β IEEE Open Journal of Circuits and Systems 2: 648β59.
Shin, Richard, and Dawn Song. 2017. βJPEG-Resistant Adversarial Images.β In NIPS 2017 Workshop on Machine Learning and Computer Security, 1:8.
Toderici, George, Sean M. OβMalley, Sung Jin Hwang, Damien Vincent, David Minnen, Shumeet Baluja, Michele Covell, and Rahul Sukthankar. 2016. βVariable Rate Image Compression with Recurrent Neural Networks.β arXiv.
No comments yet. Why not leave one?