Parameter Training Efficiency Aware Resource Allocation for AIGC in Space-Air-Ground Integrated Networks
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909979312652288 |
|---|---|
| author | Qian, Liangxin Si, Peiyuan Zhao, Jun Lam, Kwok-Yan |
| author_facet | Qian, Liangxin Si, Peiyuan Zhao, Jun Lam, Kwok-Yan |
| contents | With the evolution of artificial intelligence-generated content (AIGC) techniques and the development of space-air-ground integrated networks (SAGIN), there will be a growing opportunity to enhance more users' mobile experience with customized AIGC applications. This is made possible through the use of parameter-efficient fine-tuning (PEFT) training alongside mobile edge computing. In this paper, we formulate the optimization problem of maximizing the parameter training efficiency of the SAGIN system over wireless networks under limited resource constraints. We propose the Parameter training efficiency Aware Resource Allocation (PARA) technique to jointly optimize user association, data offloading, and communication and computational resource allocation. Solid proofs are presented to solve this difficult sum of ratios problem based on quadratically constrained quadratic programming (QCQP), semidefinite programming (SDP), graph theory, and fractional programming (FP) techniques. Our proposed PARA technique is effective in finding a stationary point of this non-convex problem. The simulation results demonstrate that the proposed PARA method outperforms other baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_13602 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Parameter Training Efficiency Aware Resource Allocation for AIGC in Space-Air-Ground Integrated Networks Qian, Liangxin Si, Peiyuan Zhao, Jun Lam, Kwok-Yan Emerging Technologies Signal Processing With the evolution of artificial intelligence-generated content (AIGC) techniques and the development of space-air-ground integrated networks (SAGIN), there will be a growing opportunity to enhance more users' mobile experience with customized AIGC applications. This is made possible through the use of parameter-efficient fine-tuning (PEFT) training alongside mobile edge computing. In this paper, we formulate the optimization problem of maximizing the parameter training efficiency of the SAGIN system over wireless networks under limited resource constraints. We propose the Parameter training efficiency Aware Resource Allocation (PARA) technique to jointly optimize user association, data offloading, and communication and computational resource allocation. Solid proofs are presented to solve this difficult sum of ratios problem based on quadratically constrained quadratic programming (QCQP), semidefinite programming (SDP), graph theory, and fractional programming (FP) techniques. Our proposed PARA technique is effective in finding a stationary point of this non-convex problem. The simulation results demonstrate that the proposed PARA method outperforms other baselines. |
| title | Parameter Training Efficiency Aware Resource Allocation for AIGC in Space-Air-Ground Integrated Networks |
| topic | Emerging Technologies Signal Processing |
| url | https://arxiv.org/abs/2406.13602 |