# DeepIoT: Compressing Deep Neural Network Structures for Sensing Systems with a Compressor-Critic Framework

arXiv: https://arxiv.org/abs/1706.01215 · HTML: https://mlsanity.com/paper/1706.01215/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Jun 2017.

## Built on by (most cited first, 22 of 22)

- [Machine Learning for Microcontroller-Class Hardware: A Review](https://mlsanity.com/paper/2205.14550/) (2022)
- [Meta-HAR: Federated Representation Learning for Human Activity Recognition](https://mlsanity.com/paper/2106.00615/) (2021)
- [Designing and Training of Lightweight Neural Networks on Edge Devices using Early Halting in Knowledge Distillation](https://mlsanity.com/paper/2209.15560/) (2022)
- [Compacting Deep Neural Networks for Internet of Things: Methods and Applications](https://mlsanity.com/paper/2103.11083/) (2021)
- [Boosting Distributed Machine Learning Training Through Loss-tolerant Transmission Protocol](https://mlsanity.com/paper/2305.04279/) (2023)
- [P3SL: Personalized Privacy-Preserving Split Learning on Heterogeneous Edge Devices](https://mlsanity.com/paper/2507.17228/) (2025)
- [Physical Self-Supervised Learning: IMU Sensing without Manual Labels](https://mlsanity.com/paper/2607.18361/) (2026)
- [AdaSpring: Context-adaptive and Runtime-evolutionary Deep Model Compression for Mobile Applications](https://mlsanity.com/paper/2101.11800/) (2021)
- [SmartDet: Context-Aware Dynamic Control of Edge Task Offloading for Mobile Object Detection](https://mlsanity.com/paper/2201.04235/) (2022)
- [The Effects of Partitioning Strategies on Energy Consumption in Distributed CNN Inference at The Edge](https://mlsanity.com/paper/2210.08392/) (2022)
- [Edge-MultiAI: Multi-Tenancy of Latency-Sensitive Deep Learning Applications on Edge](https://mlsanity.com/paper/2211.07130/) (2022)
- [A Converting Autoencoder Toward Low-latency and Energy-efficient DNN Inference at the Edge](https://mlsanity.com/paper/2403.07036/) (2024)
- [MOHAQ: Multi-Objective Hardware-Aware Quantization of Recurrent Neural Networks](https://mlsanity.com/paper/2108.01192/) (2021)
- [FCN-Pose: A Pruned and Quantized CNN for Robot Pose Estimation for Constrained Devices](https://mlsanity.com/paper/2205.13272/) (2022)
- [A Review of the Convergence of 5G/6G Architecture and Deep Learning](https://mlsanity.com/paper/2208.07643/) (2022)
- [Mobiprox: Supporting Dynamic Approximate Computing on Mobiles](https://mlsanity.com/paper/2303.11291/) (2023)
- [Manifold Regularization for Memory-Efficient Training of Deep Neural Networks](https://mlsanity.com/paper/2305.17119/) (2023)
- [Memory-efficient Energy-adaptive Inference of Pre-Trained Models on Batteryless Embedded Systems](https://mlsanity.com/paper/2405.10426/) (2024)
- [CSRAP: Enhanced Canvas Attention Scheduling for Real-Time Mission Critical Perception](https://mlsanity.com/paper/2508.04976/) (2025)
- [Fast and accurate neural reflectance transformation imaging through knowledge distillation](https://mlsanity.com/paper/2510.24486/) (2025)
- [Efficient-Husformer: Efficient Multimodal Transformer Hyperparameter Optimization for Stress and Cognitive Loads](https://mlsanity.com/paper/2511.22362/) (2025)
- [LeanStream: A Speculate-and-Refine Streaming Framework for Efficient on-Device LLM Inference](https://mlsanity.com/paper/2609.03079/) (2026)

## Abstract (arXiv)

Recent advances in deep learning motivate the use of deep neutral networks in sensing applications, but their excessive resource needs on constrained embedded devices remain an important impediment. A recently explored solution space lies in compressing (approximating or simplifying) deep neural networks in some manner before use on the device. We propose a new compression solution, called DeepIoT, that makes two key contributions in that space. First, unlike current solutions geared for compressing specific types of neural networks, DeepIoT presents a unified approach that compresses all commonly used deep learning structures for sensing applications, including fully-connected, convolutional, and recurrent neural networks, as well as their combinations. Second, unlike solutions that either sparsify weight matrices or assume linear structure within weight matrices, DeepIoT compresses neural network structures into smaller dense matrices by finding the minimum number of non-redundant hidden elements, such as filters and dimensions required by each layer, while keeping the performance of sensing applications the same. Importantly, it does so using an approach that obtains a global view of parameter redundancies, which is shown to produce superior compression. We conduct experiments with five different sensing-related tasks on Intel Edison devices. DeepIoT outperforms all compared baseline algorithms with respect to execution time and energy consumption by a significant margin. It reduces the size of deep neural networks by 90% to 98.9%. It is thus able to shorten execution time by 71.4% to 94.5%, and decrease energy consumption by 72.2% to 95.7%. These improvements are achieved without loss of accuracy. The results underscore the potential of DeepIoT for advancing the exploitation of deep neural networks on resource-constrained embedded devices.