# Edge AI: Evaluation of Model Compression Techniques for Convolutional Neural Networks

arXiv: https://arxiv.org/abs/2409.02134 · HTML: https://mlsanity.com/paper/2409.02134/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 58 references that are in the map.
- First posted on arXiv 2 Sep 2024.

## Builds on (most cited first, 22 of 22)

- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Pruning Filters for Efficient ConvNets](https://mlsanity.com/paper/1608.08710/) (2016)
- [SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size](https://mlsanity.com/paper/1602.07360/) (2016)
- [Do Deep Nets Really Need to be Deep?](https://mlsanity.com/paper/1312.6184/) (2013)
- [Paraphrasing Complex Network: Network Compression via Factor Transfer](https://mlsanity.com/paper/1802.04977/) (2018)
- [Bayesian Compression for Deep Learning](https://mlsanity.com/paper/1705.08665/) (2017)
- [Wide Compression: Tensor Ring Nets](https://mlsanity.com/paper/1802.09052/) (2018)
- [OTOV2: Automatic, Generic, User-Friendly](https://mlsanity.com/paper/2303.06862/) (2023)
- [Deep Neural Network Compression with Single and Multiple Level Quantization](https://mlsanity.com/paper/1803.03289/) (2018)
- [Coreset-Based Neural Network Compression](https://mlsanity.com/paper/1807.09810/) (2018)
- [HALOC: Hardware-Aware Automatic Low-Rank Compression for Compact Neural Networks](https://mlsanity.com/paper/2301.09422/) (2023)
- [Geometry-aware training of factorized layers in tensor Tucker format](https://mlsanity.com/paper/2305.19059/) (2023)
- [DeepSZ: A Novel Framework to Compress Deep Neural Networks by Using Error-Bounded Lossy Compression](https://mlsanity.com/paper/1901.09124/) (2019)
- [Joint Matrix Decomposition for Deep Convolutional Neural Networks Compression](https://mlsanity.com/paper/2107.04386/) (2021)
- [ADA-Tucker: Compressing Deep Neural Networks via Adaptive Dimension Adjustment Tucker Decomposition](https://mlsanity.com/paper/1906.07671/) (2019)
- [Deep Learning at the Edge](https://mlsanity.com/paper/1910.10231/) (2019)
- [Compressing Neural Networks Using Tensor Networks with Exponentially Fewer Variational Parameters](https://mlsanity.com/paper/2305.06058/) (2023)
- [Teacher-Student Compression with Generative Adversarial Networks](https://mlsanity.com/paper/1812.02271/) (2018)
- [Recovering from Random Pruning: On the Plasticity of Deep Convolutional Neural Networks](https://mlsanity.com/paper/1801.10447/) (2018)
- [OTOv3: Automatic Architecture-Agnostic Neural Network Training and Compression from Structured Pruning to Erasing Operators](https://mlsanity.com/paper/2312.09411/) (2023)
- [Lossy and Lossless (L2) Post-training Model Size Compression](https://mlsanity.com/paper/2308.04269/) (2023)
- [Filter Pruning via Filters Similarity in Consecutive Layers](https://mlsanity.com/paper/2304.13397/) (2023)

## Built on by (most cited first, 4 of 4)

- [Advanced Knowledge Transfer: Refined Feature Distillation for Zero-Shot Quantization in Edge Computing](https://mlsanity.com/paper/2412.19125/) (2024)
- [QIANets: Quantum-Integrated Adaptive Networks for Reduced Latency and Improved Inference Times in CNN Models](https://mlsanity.com/paper/2410.10318/) (2024)
- [maxVSTAR: Maximally Adaptive Vision-Guided CSI Sensing with Closed-Loop Edge Model Adaptation for Robust Human Activity Recognition](https://mlsanity.com/paper/2510.26146/) (2025)
- [A Comparative Study of CNN Optimization Methods for Edge AI: Exploring the Role of Early Exits](https://mlsanity.com/paper/2604.14789/) (2026)

## Abstract (arXiv)

This work evaluates the compression techniques on ConvNeXt models in image classification tasks using the CIFAR-10 dataset. Structured pruning, unstructured pruning, and dynamic quantization methods are evaluated to reduce model size and computational complexity while maintaining accuracy. The experiments, conducted on cloud-based platforms and edge device, assess the performance of these techniques. Results show significant reductions in model size, with up to 75% reduction achieved using structured pruning techniques. Additionally, dynamic quantization achieves a reduction of up to 95% in the number of parameters. Fine-tuned models exhibit improved compression performance, indicating the benefits of pre-training in conjunction with compression techniques. Unstructured pruning methods reveal trends in accuracy and compression, with limited reductions in computational complexity. The combination of OTOV3 pruning and dynamic quantization further enhances compression performance, resulting 89.7% reduction in size, 95% reduction with number of parameters and MACs, and 3.8% increase with accuracy. The deployment of the final compressed model on edge device demonstrates high accuracy 92.5% and low inference time 20 ms, validating the effectiveness of compression techniques for real-world edge computing applications.