# A Two-Stage Data Selection Framework for Data-Efficient Model Training on Edge Devices

arXiv: https://arxiv.org/abs/2505.16563 · HTML: https://mlsanity.com/paper/2505.16563/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 68 references that are in the map.
- First posted on arXiv 22 May 2025.

## Builds on (most cited first, 27 of 27)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size](https://mlsanity.com/paper/1602.07360/) (2016)
- [Revisiting Unreasonable Effectiveness of Data in Deep Learning Era](https://mlsanity.com/paper/1707.02968/) (2017)
- [Towards Federated Learning at Scale: System Design](https://mlsanity.com/paper/1902.01046/) (2019)
- [Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition](https://mlsanity.com/paper/1804.03209/) (2018)
- [Data Shapley: Equitable Valuation of Data for Machine Learning](https://mlsanity.com/paper/1904.02868/) (2019)
- [Selection via Proxy: Efficient Data Selection for Deep Learning](https://mlsanity.com/paper/1906.11829/) (2019)
- [Not All Samples Are Created Equal: Deep Learning with Importance Sampling](https://mlsanity.com/paper/1803.00942/) (2018)
- [Don't Decay the Learning Rate, Increase the Batch Size](https://mlsanity.com/paper/1711.00489/) (2017)
- [Stochastic Optimization with Importance Sampling](https://mlsanity.com/paper/1401.2753/) (2014)
- [Adaptive Second Order Coresets for Data-efficient Machine Learning](https://mlsanity.com/paper/2207.13887/) (2022)
- [Pixie: A System for Recommending 3+ Billion Items to 200+ Million Users in Real-Time](https://mlsanity.com/paper/1711.07601/) (2017)
- [A First Look at Deep Learning Apps on Smartphones](https://mlsanity.com/paper/1812.05448/) (2018)
- [Biased Importance Sampling for Deep Neural Network Training](https://mlsanity.com/paper/1706.00043/) (2017)
- [Mandheling: Mixed-Precision On-Device DNN Training with DSP Offloading](https://mlsanity.com/paper/2206.07509/) (2022)
- [Choosing the Sample with Lowest Loss makes SGD Robust](https://mlsanity.com/paper/2001.03316/) (2020)
- [GIO: Gradient Information Optimization for Training Dataset Selection](https://mlsanity.com/paper/2306.11670/) (2023)
- [ElasticTrainer: Speeding Up On-Device Training with Runtime Elastic Tensor Selection](https://mlsanity.com/paper/2312.14227/) (2023)
- [SurveilEdge: Real-time Video Query based on Collaborative Cloud-Edge Deep Learning](https://mlsanity.com/paper/2001.01043/) (2020)
- [To Store or Not? Online Data Selection for Federated Learning with Limited Storage](https://mlsanity.com/paper/2209.00195/) (2022)
- [Boosting DNN Cold Inference on Edge Devices](https://mlsanity.com/paper/2206.07446/) (2022)
- [Enabling On-Device Smartphone GPU based Training: Lessons Learned](https://mlsanity.com/paper/2202.10100/) (2022)
- [FedSS: Federated Learning with Smart Selection of clients](https://mlsanity.com/paper/2207.04569/) (2022)

## Built on by (most cited first, 2 of 2)

- [LearnAlign: Data Selection for LLM Reinforcement Learning with Improved Gradient Alignment](https://mlsanity.com/paper/2506.11480/) (2025)
- [Optimizing Storage Overhead of User Behavior Log for ML-embedded Mobile Apps](https://mlsanity.com/paper/2510.13405/) (2025)

## Abstract (arXiv)

The demand for machine learning (ML) model training on edge devices is escalating due to data privacy and personalized service needs. However, we observe that current on-device model training is hampered by the under-utilization of on-device data, due to low training throughput, limited storage and diverse data importance. To improve data resource utilization, we propose a two-stage data selection framework sf Titan to select the most important data batch from streaming data for model training with guaranteed efficiency and effectiveness. Specifically, in the first stage, sf Titan filters out a candidate dataset with potentially high importance in a coarse-grained manner.In the second stage of fine-grained selection, we propose a theoretically optimal data selection strategy to identify the data batch with the highest model performance improvement to current training round. To further enhance time-and-resource efficiency, sf Titan leverages a pipeline to co-execute data selection and model training, and avoids resource conflicts by exploiting idle computing resources. We evaluate sf Titan on real-world edge devices and three representative edge computing tasks with diverse models and data modalities. Empirical results demonstrate that sf Titan achieves up to 43% reduction in training time and 6.2% increase in final accuracy with minor system overhead, such as data processing delay, memory footprint and energy consumption.