LAMBO: Large AI Model Empowered Edge Intelligence

doi:10.1109/MCOM.001.2400076

LAMBO: Large AI Model Empowered Edge Intelligence

2024-01-01

0

OA

AI

DOI：10.1109/MCOM.001.2400076

原文链接

原文求助

摘要

En 中文

Next-generation edge intelligence is anticipated to benefit various applications via offloading techniques. However, traditional offloading architectures face several issues, including heterogeneous constraints, partial perception, uncertain generalization, and lack of tractability. In this article, we propose a large AI model-based offloading (LAMBO) framework with over one billion parameters for solving these problems. We first use input embedding (IE) to achieve normalized feature representation with heterogeneous constraints and task prompts. Then, we introduce a novel asymmetric encoder-decoder (AED) as the decision-making model, which is an improved transformer architecture consisting of a deep encoder and a shallow decoder for global perception and decision. Next, actor-critic learning (ACL) is used to pre-train the AED for different optimization tasks under corresponding prompts, enhancing the AED's generalization in multi-task scenarios. Finally, we propose an active learning from expert feedback (ALEF) method to fine-tune the decoder of the AED for tracking changes in dynamic environments. Our simulation results validate the advantages of the proposed LAMBO framework.

Keyword:

Training

Decision making

Computer architecture

Optimization

Transformers

Decoding

Data models

Deep learning

Adaptation models

Reinforcement learning

LAMBO: Large AI Model Empowered Edge Intelligence

摘要

期刊

机构

引用论文