Lossless and Near-Lossless Compression for Foundation Models

• URL: http://arxiv.org/abs/2404.15198v1
• Date: Fri, 5 Apr 2024 16:52:55 GMT
• Title: Lossless and Near-Lossless Compression for Foundation Models
• Authors: Moshik Hershcovitch, Leshem Choshen, Andrew Wood, Ilias Enmouri, Peter Chin, Swaminathan Sundararaman, Danny Harnik,
• Abstract summary: We investigate the source of model compressibility, introduce compression variants tailored for models and categorize models to compressibility groups. We estimate that these methods could save over an ExaByte per month of network traffic downloaded from a large model hub like HuggingFace.
• Score: 11.307357041746865
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: With the growth of model sizes and scale of their deployment, their sheer size burdens the infrastructure requiring more network and more storage to accommodate these. While there is a vast literature about reducing model sizes, we investigate a more traditional type of compression -- one that compresses the model to a smaller form and is coupled with a decompression algorithm that returns it to its original size -- namely lossless compression. Somewhat surprisingly, we show that such lossless compression can gain significant network and storage reduction on popular models, at times reducing over $50\%$ of the model size. We investigate the source of model compressibility, introduce compression variants tailored for models and categorize models to compressibility groups. We also introduce a tunable lossy compression technique that can further reduce size even on the less compressible models with little to no effect on the model accuracy. We estimate that these methods could save over an ExaByte per month of network traffic downloaded from a large model hub like HuggingFace.

Related papers

• ZipNN: Lossless Compression for AI Models [10.111136691015554]
  We present ZipNN a lossless compression tailored to neural networks. On popular models (e.g. Llama 3) ZipNN shows space savings that are over 17% better than vanilla compression. We estimate that these methods could save over an ExaByte per month of network traffic downloaded from a large model hub like Hugging Face.
  arXiv  Detail & Related papers  (2024-11-07T23:28:23Z)
• MoDeGPT: Modular Decomposition for Large Language Model Compression [59.361006801465344]
  This paper introduces textbfModular bfDecomposition (MoDeGPT), a novel structured compression framework. MoDeGPT partitions the Transformer block into modules comprised of matrix pairs and reduces the hidden dimensions. Our experiments show MoDeGPT, without backward propagation, matches or surpasses previous structured compression methods.
  arXiv  Detail & Related papers  (2024-08-19T01:30:14Z)
• Tiny Models are the Computational Saver for Large Models [1.8350044465969415]
  This paper introduces TinySaver, an early-exit-like dynamic model compression approach which employs tiny models to substitute large models adaptively. Our evaluation of this approach in ImageNet-1k classification demonstrates its potential to reduce the number of compute operations by up to 90%, with only negligible losses in performance.
  arXiv  Detail & Related papers  (2024-03-26T14:14:30Z)
• Activations and Gradients Compression for Model-Parallel Training [85.99744701008802]
  We study how simultaneous compression of activations and gradients in model-parallel distributed training setup affects convergence. We find that gradients require milder compression rates than activations. Experiments also show that models trained with TopK perform well only when compression is also applied during inference.
  arXiv  Detail & Related papers  (2024-01-15T15:54:54Z)
• Lossy and Lossless (L$^2$) Post-training Model Size Compression [12.926354646945397]
  We propose a post-training model size compression method that combines lossy and lossless compression in a unified way. Our method can achieve a stable $10times$ compression ratio without sacrificing accuracy and a $20times$ compression ratio with minor accuracy loss in a short time.
  arXiv  Detail & Related papers  (2023-08-08T14:10:16Z)
• Estimating the Resize Parameter in End-to-end Learned Image Compression [50.20567320015102]
  We describe a search-free resizing framework that can further improve the rate-distortion tradeoff of recent learned image compression models. Our results show that our new resizing parameter estimation framework can provide Bjontegaard-Delta rate (BD-rate) improvement of about 10% against leading perceptual quality engines.
  arXiv  Detail & Related papers  (2022-04-26T01:35:02Z)
• What do Compressed Large Language Models Forget? Robustness Challenges in Model Compression [68.82486784654817]
  We study two popular model compression techniques including knowledge distillation and pruning. We show that compressed models are significantly less robust than their PLM counterparts on adversarial test sets. We develop a regularization strategy for model compression based on sample uncertainty.
  arXiv  Detail & Related papers  (2021-10-16T00:20:04Z)
• Towards Compact CNNs via Collaborative Compression [166.86915086497433]
  We propose a Collaborative Compression scheme, which joints channel pruning and tensor decomposition to compress CNN models. We achieve 52.9% FLOPs reduction by removing 48.4% parameters on ResNet-50 with only a Top-1 accuracy drop of 0.56% on ImageNet 2012.
  arXiv  Detail & Related papers  (2021-05-24T12:07:38Z)
• Lossless Compression with Latent Variable Models [4.289574109162585]
  We use latent variable models, which we call 'bits back with asymmetric numeral systems' (BB-ANS) The method involves interleaving encode and decode steps, and achieves an optimal rate when compressing batches of data. We describe 'Craystack', a modular software framework which we have developed for rapid prototyping of compression using deep generative models.
  arXiv  Detail & Related papers  (2021-04-21T14:03:05Z)
• GAN Slimming: All-in-One GAN Compression by A Unified Optimization Framework [94.26938614206689]
  We propose the first unified optimization framework combining multiple compression means for GAN compression, dubbed GAN Slimming. We apply GS to compress CartoonGAN, a state-of-the-art style transfer network, by up to 47 times, with minimal visual quality degradation.
  arXiv  Detail & Related papers  (2020-08-25T14:39:42Z)
• Self-Supervised GAN Compression [32.21713098893454]
  We show that a standard model compression technique, weight pruning, cannot be applied to GANs using existing methods. We then develop a self-supervised compression technique which uses the trained discriminator to supervise the training of a compressed generator. We show that this framework has a compelling performance to high degrees of sparsity, can be easily applied to new tasks and models, and enables meaningful comparisons between different pruning granularities.
  arXiv  Detail & Related papers  (2020-07-03T04:18:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.