# Evaluating Post-Training Compression in GANs using Locality-Sensitive Hashing

arXiv: https://arxiv.org/abs/2103.11912 · HTML: https://mlsanity.com/paper/2103.11912/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 39 references that are in the map.
- First posted on arXiv 22 Mar 2021.

## Builds on (most cited first, 27 of 27)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [Spectral Normalization for Generative Adversarial Networks](https://mlsanity.com/paper/1802.05957/) (2018)
- [Demystifying MMD GANs](https://mlsanity.com/paper/1801.01401/) (2018)
- [Training Generative Adversarial Networks with Limited Data](https://mlsanity.com/paper/2006.06676/) (2020)
- [StarGAN v2: Diverse Image Synthesis for Multiple Domains](https://mlsanity.com/paper/1912.01865/) (2019)
- [Improved Precision and Recall Metric for Assessing Generative Models](https://mlsanity.com/paper/1904.06991/) (2019)
- [Assessing Generative Models via Precision and Recall](https://mlsanity.com/paper/1806.00035/) (2018)
- [Improving Neural Network Quantization without Retraining using Outlier Channel Splitting](https://mlsanity.com/paper/1901.09504/) (2019)
- [Self-Supervised GANs via Auxiliary Rotation Loss](https://mlsanity.com/paper/1811.11212/) (2018)
- [Improved Consistency Regularization for GANs](https://mlsanity.com/paper/2002.04724/) (2020)
- [GAN Slimming: All-in-One GAN Compression by A Unified Optimization Framework](https://mlsanity.com/paper/2008.11062/) (2020)
- [Distilling portable Generative Adversarial Networks for Image Translation](https://mlsanity.com/paper/2003.03519/) (2020)
- [Co-Evolutionary Compression for Unpaired Image Translation](https://mlsanity.com/paper/1907.10804/) (2019)
- [Compressing GANs using Knowledge Distillation](https://mlsanity.com/paper/1902.00159/) (2019)
- [QGAN: Quantized Generative Adversarial Networks](https://mlsanity.com/paper/1901.08263/) (2019)
- [Dropout-GAN: Learning from a Dynamic Ensemble of Discriminators](https://mlsanity.com/paper/1807.11346/) (2018)
- [Progressive Augmentation of GANs](https://mlsanity.com/paper/1901.10422/) (2019)
- [microbatchGAN: Stimulating Diversity with Multi-Adversarial Discrimination](https://mlsanity.com/paper/2001.03376/) (2020)
- [Self-Supervised GAN Compression](https://mlsanity.com/paper/2007.01491/) (2020)
- [Mark-Evaluate: Assessing Language Generation using Population Estimation Methods](https://mlsanity.com/paper/2010.04606/) (2020)
- [Instant Quantization of Neural Networks using Monte Carlo Methods](https://mlsanity.com/paper/1905.12253/) (2019)
- [PBGen: Partial Binarization of Deconvolution-Based Generators for Edge Intelligence](https://mlsanity.com/paper/1802.09153/) (2018)

## Built on by (most cited first, 1 of 1)

- [Cut Inner Layers: A Structured Pruning Strategy for Efficient U-Net GANs](https://mlsanity.com/paper/2206.14658/) (2022)

## Abstract (arXiv)

The analysis of the compression effects in generative adversarial networks (GANs) after training, i.e. without any fine-tuning, remains an unstudied, albeit important, topic with the increasing trend of their computation and memory requirements. While existing works discuss the difficulty of compressing GANs during training, requiring novel methods designed with the instability of GANs training in mind, we show that existing compression methods (namely clipping and quantization) may be directly applied to compress GANs post-training, without any additional changes. High compression levels may distort the generated set, likely leading to an increase of outliers that may negatively affect the overall assessment of existing k-nearest neighbor (KNN) based metrics. We propose two new precision and recall metrics based on locality-sensitive hashing (LSH), which, on top of increasing the outlier robustness, decrease the complexity of assessing an evaluation sample against n reference samples from O(n) to O(log(n)), if using LSH and KNN, and to O(1), if only applying LSH. We show that low-bit compression of several pre-trained GANs on multiple datasets induces a trade-off between precision and recall, retaining sample quality while sacrificing sample diversity.