paper-with-me

Papers

DC-CCL: Device-Cloud Collaborative Controlled Learning for Large Vision Models

2023-03-18 · Yucheng Ding, Chaoyue Niu, Fan Wu, Shaojie Tang, Chengfei Lyu, Guihai Chen

Many large vision models have been deployed on the cloud for real-time services. Meanwhile, fresh samples are continuously generated on the served mobile device. How to leverage the device-side samples to improve the cloud-side large model becomes a practical requirement, but falls into the dilemma of no raw sample up-link and no large model down-link. Specifically, the user may opt out of sharing raw samples with the cloud due to the concern of privacy or communication overhead, while the size of some large vision models far exceeds the mobile device's runtime capacity. In this work, we propose a device-cloud collaborative controlled learning framework, called DC-CCL, enabling a cloud-side large vision model that cannot be directly deployed on the mobile device to still benefit from the device-side local samples. In particular, DC-CCL vertically splits the base model into two submodels, one large submodel for learning from the cloud-side samples and the other small submodel for learning from the device-side samples and performing device-cloud knowledge fusion. Nevertheless, on-device training of the small submodel requires the output of the cloud-side large submodel to compute the desired gradients. DC-CCL thus introduces a light-weight model to mimic the large cloud-side submodel with knowledge distillation, which can be offloaded to the mobile device to control its small submodel's optimization direction. Given the decoupling nature of two submodels in collaborative learning, DC-CCL also allows the cloud to take a pre-trained model and the mobile device to take another model with a different backbone architecture.

📄 PDF Abstract BibTeX arXiv:2303.10361

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge Distillation

Methods 이 논문이 사용한 방법론

OPT OPT is a suite of decoder-only pre-trained transformers ranging from 125M to 175B parameters. The model uses an AdamW optimizer and weight decay of 0.1. It follows a linear…
BASE 설명 없음

Similar Papers 제목 키워드 기반

Janus: Collaborative Vision Transformer Under Dynamic Network Environment

2025-02-14 · Linyi Jiang, Silvery D. Fu, Yifei Zhu, Bo Li

Vision Transformers (ViTs) have outperformed traditional Convolutional Neural Network architectures and achieved state-of-the-art results in various computer vision tasks. Since ViTs are computationally expensive, the mo…

ECLM: Efficient Edge-Cloud Collaborative Learning with Continuous Environment Adaptation

2023-11-18 · Yan Zhuang, Zhenzhe Zheng, Yunfeng Shao, Bingshuai Li 외

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, …

Hybrid SD: Edge-Cloud Collaborative Inference for Stable Diffusion Models

2024-08-13 · Chenqian Yan, Songwei Liu, Hongjian Liu, Xurui Peng 외

Stable Diffusion Models (SDMs) have shown remarkable proficiency in image synthesis. However, their broad application is impeded by their large model sizes and intensive computational requirements, which typically requir…

Collaborative InferenceDiversityImage Generation

Cloud-Device Collaborative Learning for Multimodal Large Language Models

2023-12-26 · CVPR 2024 1 · Guanqun Wang, Jiaming Liu, Chenxuan Li, Junpeng Ma 외

The burgeoning field of Multimodal Large Language Models (MLLMs) has exhibited remarkable performance in diverse tasks such as captioning, commonsense reasoning, and visual scene understanding. However, the deployment of…

Device-Cloud CollaborationKnowledge DistillationScene Understanding

Collaboration of Large Language Models and Small Recommendation Models for Device-Cloud Recommendation

2025-01-10 · Zheqi Lv, Tianyu Zhan, Wenjie Wang, Xinyu Lin 외

Large Language Models (LLMs) for Recommendation (LLM4Rec) is a promising research direction that has demonstrated exceptional performance in this field. However, its inability to capture real-time user preferences greatl…

Collaborative InferenceDevice-Cloud CollaborationEdge-computingReranking