# Continual Learning at the Edge: Real-Time Training on Smartphone Devices

arXiv: https://arxiv.org/abs/2105.13127 · HTML: https://mlsanity.com/paper/2105.13127/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 8 references that are in the map.
- First posted on arXiv 24 May 2021.

## Builds on (most cited first, 7 of 7)

- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [CORe50: a New Dataset and Benchmark for Continuous Object Recognition](https://mlsanity.com/paper/1705.03550/) (2017)
- [Avalanche: an End-to-End Library for Continual Learning](https://mlsanity.com/paper/2104.00405/) (2021)
- [Latent Replay for Real-Time Continual Learning](https://mlsanity.com/paper/1912.01100/) (2019)
- [Replay in Deep Learning: Current Approaches and Missing Biological Elements](https://mlsanity.com/paper/2104.04132/) (2021)
- [Rehearsal-Free Continual Learning over Small Non-I.I.D. Batches](https://mlsanity.com/paper/1907.03799/) (2019)
- [Continual Learning on the Edge with TensorFlow Lite](https://mlsanity.com/paper/2105.01946/) (2021)

## Built on by (most cited first, 20 of 20)

- [When Foundation Model Meets Federated Learning: Motivations, Challenges, and Future Directions](https://mlsanity.com/paper/2306.15546/) (2023)
- [SparCL: Sparse Continual Learning on the Edge](https://mlsanity.com/paper/2209.09476/) (2022)
- [Online Continual Learning for Embedded Devices](https://mlsanity.com/paper/2203.10681/) (2022)
- [One-stage Prompt-based Continual Learning](https://mlsanity.com/paper/2402.16189/) (2024)
- [Continual Learning in the Frequency Domain](https://mlsanity.com/paper/2410.06645/) (2024)
- [Cost-effective On-device Continual Learning over Memory Hierarchy with Miro](https://mlsanity.com/paper/2308.06053/) (2023)
- [Online Continual Learning for Robust Indoor Object Recognition](https://mlsanity.com/paper/2307.09827/) (2023)
- [Online Continual Learning in Keyword Spotting for Low-Resource Devices via Pooling High-Order Temporal Statistics](https://mlsanity.com/paper/2307.12660/) (2023)
- [When Continue Learning Meets Multimodal Large Language Model: A Survey](https://mlsanity.com/paper/2503.01887/) (2025)
- [Generative Negative Replay for Continual Learning](https://mlsanity.com/paper/2204.05842/) (2022)
- [Exploring the Stability Gap in Continual Learning: The Role of the Classification Head](https://mlsanity.com/paper/2411.04723/) (2024)
- [12 mJ per Class On-Device Online Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2403.07851/) (2024)
- [EdgeOL: Efficient in-situ Online Learning on Edge Devices](https://mlsanity.com/paper/2401.16694/) (2024)
- [Increasing Depth of Neural Networks for Life-long Learning](https://mlsanity.com/paper/2202.10821/) (2022)
- [Efficient Continual Learning through Frequency Decomposition and Integration](https://mlsanity.com/paper/2503.22175/) (2025)
- [Federated Learning -- Methods, Applications and beyond](https://mlsanity.com/paper/2212.11729/) (2022)
- [Data-Free Generative Replay for Class-Incremental Learning on Imbalanced Data](https://mlsanity.com/paper/2406.09052/) (2024)
- [Online Continual Learning on Intel Loihi 2 via a Co-designed Spiking Neural Network](https://mlsanity.com/paper/2511.01553/) (2025)
- [Critical Patch-Aware Sparse Prompting with Decoupled Training for Continual Learning on the Edge](https://mlsanity.com/paper/2604.07399/) (2026)
- [HEDP: A Hybrid Energy-Distance Prompt-based Framework for Domain Incremental Learning](https://mlsanity.com/paper/2605.05776/) (2026)

## Abstract (arXiv)

On-device training for personalized learning is a challenging research problem. Being able to quickly adapt deep prediction models at the edge is necessary to better suit personal user needs. However, adaptation on the edge poses some questions on both the efficiency and sustainability of the learning process and on the ability to work under shifting data distributions. Indeed, naively fine-tuning a prediction model only on the newly available data results in catastrophic forgetting, a sudden erasure of previously acquired knowledge. In this paper, we detail the implementation and deployment of a hybrid continual learning strategy (AR1*) on a native Android application for real-time on-device personalization without forgetting. Our benchmark, based on an extension of the CORe50 dataset, shows the efficiency and effectiveness of our solution.