ECLM: Efficient Edge-Cloud Collaborative Learning with Continuous Environment Adaptation
Not on the map. Why?
Abstract (arXiv)
Pervasive mobile AI applications primarily employ one of the two learning paradigms: cloud-based learning (with powerful large models) or on-device learning (with lightweight small models). Despite their own advantages, neither paradigm can effectively handle dynamic edge environments with frequent data distribution shifts and on-device resource fluctuations, inevitably suffering from performance degradation. In this paper, we propose ECLM, an edge-cloud collaborative learning framework for rapid model adaptation for dynamic edge environments. We first propose a novel block-level model decomposition design to decompose the original large cloud model into multiple combinable modules. By flexibly combining a subset of the modules, this design enables the derivation of compact, task-specific sub-models for heterogeneous edge devices from the large cloud model, and the seamless integration of new knowledge learned on these devices into the cloud model periodically. As such, ECLM ensures that the cloud model always provides up-to-date sub-models for edge devices. We further propose an end-to-end learning framework that incorporates the modular model design into an efficient model adaptation pipeline including an offline on-cloud model prototyping and training stage, and an online edge-cloud collaborative adaptation stage. Extensive experiments over various datasets demonstrate that ECLM significantly improves model performance (e.g., 18.89% accuracy increase) and resource efficiency (e.g., 7.12x communication cost reduction) in adapting models to dynamic edge environments by efficiently collaborating the edge and the cloud models.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 48 of its 75 references that are in the map. First posted on arXiv 18 Nov 2023.
Builds on
48 of 75 references.
References · 48
- 2017Attention Is All You Needcited by 59,932
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2015Distilling the Knowledge in a Neural Networkcited by 10,076
- 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
- 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
- 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
- 2016Learning without Forgettingcited by 2,465
- 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,291
- 2019Searching for MobileNetV3cited by 1,984
- 2019Federated Learning: Challenges, Methods, and Future Directionscited by 1,531
- 2018ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Designcited by 1,132
36 more
- 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,127
- 2018Federated Learning with Non-IID Datacited by 1,099
- 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 741
- 2020Ensemble Distillation for Robust Model Fusion in Federated Learningcited by 678
- 2017BranchyNet: Fast Inference via Early Exiting from Deep Neural Networkscited by 591
- 2016Learning Structured Sparsity in Deep Neural Networkscited by 538
- 2016Cross-stitch Networks for Multi-task Learningcited by 466
- 2019FedMD: Heterogenous Federated Learning via Model Distillationcited by 454
- 2018Split learning for health: Distributed deep learning without sharing raw patient datacited by 391
- 2020HeteroFL: Computation and Communication Efficient Federated Learning for Heterogeneous Clientscited by 334
- 2017SkipNet: Learning Dynamic Routing in Convolutional Networkscited by 328
- 2018Distributed learning of deep neural network over multiple agentscited by 277
- 2021FjORD: Fair and Accurate Federated Learning under heterogeneous targets with Ordered Dropoutcited by 192
- 2017BlockDrop: Dynamic Inference Paths in Residual Networkscited by 185
- 2018Expanding the Reach of Federated Learning by Reducing Client Resource Requirementscited by 162
- 2015Compression of Deep Convolutional Neural Networks for Fast and Low Power Mobile Applicationscited by 161
- 2017Adaptive Neural Networks for Efficient Inferencecited by 161
- 2018Deep Learning for Sentiment Analysis : A Surveycited by 139
- 2020Distillation-Based Semi-Supervised Federated Learning for Communication-Efficient Collaborative Training with Non-IID Private Datacited by 125
- 2020SPINN: Synergistic Progressive Inference of Neural Networks over Device and Cloudcited by 109
- 2022FedRolex: Model-Heterogeneous Federated Learning with Rolling Sub-Model Extractioncited by 102
- 2022Heterogeneous Ensemble Knowledge Transfer for Training Large Models in Federated Learningcited by 81
- 2018Adaptive Communication Strategies to Achieve the Best Error-Runtime Trade-off in Local-Update SGDcited by 76
- 2018NestDNN: Resource-Aware Multi-Tenant On-Device Deep Learning for Continuous Mobile Visioncited by 68
- 2018Modular Networks: Learning to Decompose Neural Computationcited by 62
- 2020Ekya: Continuous Learning of Video Analytics Models on Edge Compute Serverscited by 58
- 2021FedGEMS: Federated Learning of Larger Server Models via Selective Knowledge Fusioncited by 49
- 2021Edge-Cloud Polarization and Collaboration: A Comprehensive Survey for AIcited by 38
- 2020Adaptive Federated Dropout: Improving Communication Efficiency and Generalization for Federated Learningcited by 37
- 2022No One Left Behind: Inclusive Federated Learning over Heterogeneous Devicescited by 36
- 2021Device-Cloud Collaborative Learning for Recommendationcited by 34
- 2022Efficient Split-Mix Federated Learning for On-Demand and In-Situ Customizationcited by 34
- 2022GEMEL: Model Merging for Memory-Efficient, Real-Time Video Analytics at the Edgecited by 32
- 2022Walle: An End-to-End, General-Purpose, and Large-Scale Production System for Device-Cloud Collaborative Machine Learningcited by 23
- 2021LegoDNN: Block-grained Scaling of Deep Neural Networks for Mobile Visioncited by 20
- 2023AdaptiveNet: Post-deployment Neural Architecture Adaptation for Diverse Edge Environmentscited by 19
Built on by
Citing papers per year: 2025: 1.
1 paper.