# Cost-effective On-device Continual Learning over Memory Hierarchy with Miro

arXiv: https://arxiv.org/abs/2308.06053 · HTML: https://mlsanity.com/paper/2308.06053/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 66 references that are in the map.
- First posted on arXiv 11 Aug 2023.

## Builds on (most cited first, 30 of 32)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Gradient Episodic Memory for Continual Learning](https://mlsanity.com/paper/1706.08840/) (2017)
- [Continual Learning with Deep Generative Replay](https://mlsanity.com/paper/1705.08690/) (2017)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [Efficient Lifelong Learning with A-GEM](https://mlsanity.com/paper/1812.00420/) (2018)
- [Deep Networks with Stochastic Depth](https://mlsanity.com/paper/1603.09382/) (2016)
- [Dark Experience for General Continual Learning: a Strong, Simple Baseline](https://mlsanity.com/paper/2004.07211/) (2020)
- [Training Deep Nets with Sublinear Memory Cost](https://mlsanity.com/paper/1604.06174/) (2016)
- [Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence](https://mlsanity.com/paper/1801.10112/) (2018)
- [Large Scale Incremental Learning](https://mlsanity.com/paper/1905.13260/) (2019)
- [End-to-End Incremental Learning](https://mlsanity.com/paper/1807.09536/) (2018)
- [Rainbow Memory: Continual Learning with a Memory of Diverse Samples](https://mlsanity.com/paper/2103.17230/) (2021)
- [Mnemonics Training: Multi-Class Incremental Learning without Forgetting](https://mlsanity.com/paper/2002.10211/) (2020)
- [Coresets via Bilevel Optimization for Continual Learning and Streaming](https://mlsanity.com/paper/2006.03875/) (2020)
- [Deep Convolutional Neural Networks and Data Augmentation for Environmental Sound Classification](https://mlsanity.com/paper/1608.04363/) (2016)
- [GAN Memory with No Forgetting](https://mlsanity.com/paper/2006.07543/) (2020)
- [Zeus: Understanding and Optimizing GPU Energy Consumption of DNN Training](https://mlsanity.com/paper/2208.06102/) (2022)
- [Memory Replay GANs: learning to generate images from new categories without forgetting](https://mlsanity.com/paper/1809.02058/) (2018)
- [Computationally Budgeted Continual Learning: What Does Matter?](https://mlsanity.com/paper/2303.11165/) (2023)
- [Ekya: Continuous Learning of Video Analytics Models on Edge Compute Servers](https://mlsanity.com/paper/2012.10557/) (2020)
- [Continual Learning for Real-World Autonomous Systems: Algorithms, Challenges and Frameworks](https://mlsanity.com/paper/2105.12374/) (2021)
- [SparCL: Sparse Continual Learning on the Edge](https://mlsanity.com/paper/2209.09476/) (2022)
- [SpaceNet: Make Free Space For Continual Learning](https://mlsanity.com/paper/2007.07617/) (2020)
- [Continual Learning in Generative Adversarial Nets](https://mlsanity.com/paper/1705.08395/) (2017)
- [Online Continual Learning for Embedded Devices](https://mlsanity.com/paper/2203.10681/) (2022)
- [E2-Train: Training State-of-the-art CNNs with Over 80% Energy Savings](https://mlsanity.com/paper/1910.13349/) (2019)
- [Continual Learning at the Edge: Real-Time Training on Smartphone Devices](https://mlsanity.com/paper/2105.13127/) (2021)
- [Exploring System Performance of Continual Learning for Mobile and Embedded Sensing Applications](https://mlsanity.com/paper/2110.13290/) (2021)

## Built on by (most cited first, 10 of 10)

- [SwapMoE: Serving Off-the-shelf MoE-based Large Language Models with Tunable Memory Budget](https://mlsanity.com/paper/2308.15030/) (2023)
- [DaCapo: Accelerating Continuous Learning in Autonomous Systems for Video Analytics](https://mlsanity.com/paper/2403.14353/) (2024)
- [PECTP: Parameter-Efficient Cross-Task Prompts for Incremental Vision Transformer](https://mlsanity.com/paper/2407.03813/) (2024)
- [FedConv: A Learning-on-Model Paradigm for Heterogeneous Federated Clients](https://mlsanity.com/paper/2502.20639/) (2025)
- [Delta: A Cloud-assisted Data Enrichment Framework for On-Device Continual Learning](https://mlsanity.com/paper/2410.18378/) (2024)
- [EdgeOL: Efficient in-situ Online Learning on Edge Devices](https://mlsanity.com/paper/2401.16694/) (2024)
- [REP: Resource-Efficient Prompting for Rehearsal-Free Continual Learning](https://mlsanity.com/paper/2406.04772/) (2024)
- [TinyCL: An Efficient Hardware Architecture for Continual Learning on Autonomous Systems](https://mlsanity.com/paper/2402.09780/) (2024)
- [Responsive DNN Adaptation for Video Analytics against Environment Shift via Hierarchical Mobile-Cloud Collaborations](https://mlsanity.com/paper/2505.00745/) (2025)
- [SAFA-SNN: Sparsity-Aware On-Device Few-Shot Class-Incremental Learning with Fast-Adaptive Structure of Spiking Neural Network](https://mlsanity.com/paper/2510.03648/) (2025)

## Abstract (arXiv)

Continual learning (CL) trains NN models incrementally from a continuous stream of tasks. To remember previously learned knowledge, prior studies store old samples over a memory hierarchy and replay them when new tasks arrive. Edge devices that adopt CL to preserve data privacy are typically energy-sensitive and thus require high model accuracy while not compromising energy efficiency, i.e., cost-effectiveness. Our work is the first to explore the design space of hierarchical memory replay-based CL to gain insights into achieving cost-effectiveness on edge devices. We present Miro, a novel system runtime that carefully integrates our insights into the CL framework by enabling it to dynamically configure the CL system based on resource states for the best cost-effectiveness. To reach this goal, Miro also performs online profiling on parameters with clear accuracy-energy trade-offs and adapts to optimal values with low overhead. Extensive evaluations show that Miro significantly outperforms baseline systems we build for comparison, consistently achieving higher cost-effectiveness.