Live briefing
Everything about Model compression
Techniques for lossy compression of neural networks
Model compression is a machine learning technique for reducing the size of trained models. Large models can achieve high accuracy, but often at the cost of significant resource requirements. Compression techniques aim to compress models without significant performance reduction. Smaller models require less storage space, and consume less memory and compute during inference. Wikipedia →
🧭 Start here: a 6-step reading path
- Read the overview: Model compressionA 5-minute orientation
- Skim a recent paper: Model Compression for Communication Efficient Federated LearningInfluential work since 2022
- Pick up a classic book: Compression strength of composite primary structural componentsThe most-reprinted book on the subject
- Read the landmark paper: Compression model for cohesionless soilsCited 330 times
- Go sideways: Pruning (artificial neural network)A closely related topic
- Go sideways: Robust principal component analysisA closely related topic
📖 Definitions
No dictionary entry for this exact phrase.
🗂️ The subject at a glance
🏆 Landmark papers (most cited)
Compression model for cohesionless soils
A Primate Model for Chronic Nerve Compression
Private Model Compression via Knowledge Distillation
Model Compression Using Progressive Channel Pruning
Pressure Prediction Model for Compression Garment Design
Deep Model Intellectual Property Protection With Compression-Resistant Model Watermarking
A model of compression mold filling
An Artificial Compression Reduced Order Model
🆕 Influential since 2022
Model Compression for Communication Efficient Federated Learning
FedCRMW: Federated model ownership verification with compression-resistant model watermarking
Model Compression Based on Differentiable Network Channel Pruning
Analysis of Model Compression Using Knowledge Distillation
Multirate Progressive Entropy Model for Learned Image Compression
📚 Books worth reading
- 📕Compression strength of composite primary structural componentsEric R. Johnson · 1992
- 📕Near-wall modelling of compressible turbulent flowsRonald M. C. So · 1990
- 📕A near-wall four-equation turbulence model for compressible boundary layersT. P. Sommer · 1992
Amazing modelsPeter Holland · 1989- 📕Wake-shock interaction at a Mach number of 6Walsh, Michael J. · 1978
- 📕Study of two-dimensional compressible non-acoustic modeling of stirling machine type componentsRoy C. Tew · 2001
- 📕Contact stresses in pin-loaded orthotropic platesM. W. Hyer · 1984
Compressibility, Turbulence and High Speed FlowThomas B. Gatski, Jean-Paul Bonnet · 2013
Ranked by number of editions on Open Library, a rough measure of how lasting a book is.
🔗 Related topics
- Pruning (artificial neural network) In machine learning, pruning is the practice of removing parameters from an existing neural network. The goal of this process is to reduce the size (parameter…
- Robust principal component analysis Robust Principal Component Analysis (RPCA) is a modification of the widely used statistical procedure of principal component analysis (PCA) which works…
- Transformer (deep learning) In deep learning, the transformer is a family of artificial neural network architectures based on the multi-head attention mechanism, in which input data…
- Decision tree pruning Pruning is a data compression technique in machine learning and search algorithms that reduces the size of decision trees by removing sections of the tree…
- SqueezeNet SqueezeNet is a deep neural network for image classification released in 2016. SqueezeNet was developed by researchers at DeepScale, University of California…
- Weight initialization In deep learning, weight initialization or parameter initialization describes the initial step in creating a neural network. A neural network contains…
- Unsupervised learning Unsupervised learning is a framework in machine learning where, in contrast to supervised learning, algorithms learn patterns exclusively from unlabeled…
- Knowledge distillation In machine learning, knowledge distillation or model distillation is the process of transferring knowledge from a large model to a smaller one. While large…