AutoTailor: Automatic and Efficient Adaptive Model Deployment for Diverse Edge Devices
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908679147618304 |
|---|---|
| author | Liu, Mengyang Lu, Chenyu Tian, Haodong Dong, Fang Zhou, Ruiting Wang, Wei Shen, Dian Li, Guangtong Wan, Ye Li, Li |
| author_facet | Liu, Mengyang Lu, Chenyu Tian, Haodong Dong, Fang Zhou, Ruiting Wang, Wei Shen, Dian Li, Guangtong Wan, Ye Li, Li |
| contents | On-device machine learning (ML) has become a fundamental component of emerging mobile applications. Adaptive model deployment delivers efficient inference for heterogeneous device capabilities and performance requirements through customizing neural architectures. SuperNet-based approaches offer a promising solution by generating a large number of model variants from a pre-trained ML model. However, applying SuperNet in existing frameworks suffers from tedious model-aware development and time-consuming hardware-aware profiling, which limits their practical adoption.
We present AutoTailor, the first framework to enable automated, end-to-end SuperNet-based adaptive model deployment for edge devices. Unlike manual SuperNet construction, AutoTailor employs a computation graph-guided compilation approach to automatically transform user-provided ML models into SuperNets. To support efficient specialization, AutoTailor incorporates learning-free latency and accuracy predictors, enabling low-cost yet accurate performance prediction. Our extended evaluations demonstrate that AutoTailor reduces the lines of code for SuperNet construction by 11--27$\times$, decreases hardware-aware profiling costs by at least 11$\times$, and achieves up to 15.60\% absolute accuracy improvement and 60.03\% latency reduction compared to state-of-the-art approaches across diverse models and devices. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_22355 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AutoTailor: Automatic and Efficient Adaptive Model Deployment for Diverse Edge Devices Liu, Mengyang Lu, Chenyu Tian, Haodong Dong, Fang Zhou, Ruiting Wang, Wei Shen, Dian Li, Guangtong Wan, Ye Li, Li Machine Learning On-device machine learning (ML) has become a fundamental component of emerging mobile applications. Adaptive model deployment delivers efficient inference for heterogeneous device capabilities and performance requirements through customizing neural architectures. SuperNet-based approaches offer a promising solution by generating a large number of model variants from a pre-trained ML model. However, applying SuperNet in existing frameworks suffers from tedious model-aware development and time-consuming hardware-aware profiling, which limits their practical adoption. We present AutoTailor, the first framework to enable automated, end-to-end SuperNet-based adaptive model deployment for edge devices. Unlike manual SuperNet construction, AutoTailor employs a computation graph-guided compilation approach to automatically transform user-provided ML models into SuperNets. To support efficient specialization, AutoTailor incorporates learning-free latency and accuracy predictors, enabling low-cost yet accurate performance prediction. Our extended evaluations demonstrate that AutoTailor reduces the lines of code for SuperNet construction by 11--27$\times$, decreases hardware-aware profiling costs by at least 11$\times$, and achieves up to 15.60\% absolute accuracy improvement and 60.03\% latency reduction compared to state-of-the-art approaches across diverse models and devices. |
| title | AutoTailor: Automatic and Efficient Adaptive Model Deployment for Diverse Edge Devices |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2511.22355 |