# T-RECX: Tiny-Resource Efficient Convolutional neural networks with early-eXit

arXiv: https://arxiv.org/abs/2207.06613 · HTML: https://mlsanity.com/paper/2207.06613/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Early exit and edge inference, one of 12 directions in Efficient LLM inference and serving.
- Builds on 34 of its 69 references that are in the map.
- First posted on arXiv 14 Jul 2022.

## Placed in

- Efficient LLM inference and serving / [Early exit and edge inference](https://mlsanity.com/inference/early-exit-and-edge-inference/)

## Builds on (most cited first, 30 of 34)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [Learning both Weights and Connections for Efficient Neural Networks](https://mlsanity.com/paper/1506.02626/) (2015)
- [Searching for MobileNetV3](https://mlsanity.com/paper/1905.02244/) (2019)
- [How transferable are features in deep neural networks?](https://mlsanity.com/paper/1411.1792/) (2014)
- [MnasNet: Platform-Aware Neural Architecture Search for Mobile](https://mlsanity.com/paper/1807.11626/) (2018)
- [Deep Learning and the Information Bottleneck Principle](https://mlsanity.com/paper/1503.02406/) (2015)
- [Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition](https://mlsanity.com/paper/1804.03209/) (2018)
- [BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks](https://mlsanity.com/paper/1709.01686/) (2017)
- [Deeply-Supervised Nets](https://mlsanity.com/paper/1409.5185/) (2014)
- [SkipNet: Learning Dynamic Routing in Convolutional Networks](https://mlsanity.com/paper/1711.09485/) (2017)
- [Speed/accuracy trade-offs for modern convolutional object detectors](https://mlsanity.com/paper/1611.10012/) (2016)
- [Towards Understanding Knowledge Distillation](https://mlsanity.com/paper/2105.13093/) (2021)
- [Towards Optimal Structured CNN Pruning via Generative Adversarial Learning](https://mlsanity.com/paper/1903.09291/) (2019)
- [SCNN: An Accelerator for Compressed-sparse Convolutional Neural Networks](https://mlsanity.com/paper/1708.04485/) (2017)
- [MLPerf Tiny Benchmark](https://mlsanity.com/paper/2106.07597/) (2021)
- [Hello Edge: Keyword Spotting on Microcontrollers](https://mlsanity.com/paper/1711.07128/) (2017)
- [JointDNN: An Efficient Training and Inference Engine for Intelligent Mobile Cloud Computing Services](https://mlsanity.com/paper/1801.08618/) (2018)
- [Why should we add early exits to neural networks?](https://mlsanity.com/paper/2004.12814/) (2020)
- [CirCNN: Accelerating and Compressing Deep Neural Networks Using Block-CirculantWeight Matrices](https://mlsanity.com/paper/1708.08917/) (2017)
- [IDK Cascades: Fast Deep Learning by Learning not to Overthink](https://mlsanity.com/paper/1706.00885/) (2017)
- [You Look Twice: GaterNet for Dynamic Filter Selection in CNNs](https://mlsanity.com/paper/1811.11205/) (2018)
- [Visual Wake Words Dataset](https://mlsanity.com/paper/1906.05721/) (2019)
- [Compressing deep neural networks on FPGAs to binary and ternary precision with HLS4ML](https://mlsanity.com/paper/2003.06308/) (2020)
- [FastGRNN: A Fast, Accurate, Stable and Tiny Kilobyte Sized Gated Recurrent Neural Network](https://mlsanity.com/paper/1901.02358/) (2019)
- [Early Exit or Not: Resource-Efficient Blind Quality Enhancement for Compressed Images](https://mlsanity.com/paper/2006.16581/) (2020)
- [Intermittent Inference with Nonuniformly Compressed Multi-Exit Neural Network for Energy Harvesting Powered Devices](https://mlsanity.com/paper/2004.11293/) (2020)

## Built on by (most cited first, 5 of 5)

- [QUTE: Quantifying Uncertainty in TinyML with Early-exit-assisted ensembles for model-monitoring](https://mlsanity.com/paper/2404.12599/) (2024)
- [Training dynamic models using early exits for automatic speech recognition on resource-constrained devices](https://mlsanity.com/paper/2309.09546/) (2023)
- [EdgeMark: An Automation and Benchmarking System for Embedded Artificial Intelligence Tools](https://mlsanity.com/paper/2502.01700/) (2025)
- [TCUQ: Single-Pass Uncertainty Quantification from Temporal Consistency with Streaming Conformal Calibration for TinyML](https://mlsanity.com/paper/2508.12905/) (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)

Deploying Machine learning (ML) on milliwatt-scale edge devices (tinyML) is gaining popularity due to recent breakthroughs in ML and Internet of Things (IoT). Most tinyML research focuses on model compression techniques that trade accuracy (and model capacity) for compact models to fit into the KB-sized tiny-edge devices. In this paper, we show how such models can be enhanced by the addition of an early exit intermediate classifier. If the intermediate classifier exhibits sufficient confidence in its prediction, the network exits early thereby, resulting in considerable savings in time. Although early exit classifiers have been proposed in previous work, these previous proposals focus on large networks, making their techniques suboptimal/impractical for tinyML applications. Our technique is optimized specifically for tiny-CNN sized models. In addition, we present a method to alleviate the effect of network overthinking by leveraging the representations learned by the early exit. We evaluate T-RecX on three CNNs from the MLPerf tiny benchmark suite for image classification, keyword spotting and visual wake word detection tasks. Our results show that T-RecX 1) improves the accuracy of baseline network, 2) achieves 31.58% average reduction in FLOPS in exchange for one percent accuracy across all evaluated models. Furthermore, we show that our methods consistently outperform popular prior works on the tiny-CNNs we evaluate.