On-Device Dynamic DNN Inference through Spatial Sparsity Exploitation
Deep Neural Networks (DNNs) are crucial for applications like autonomous driving, augmented reality, and mixed reality. Growing concerns about latency and privacy increasingly require deploying task-specific networks on mobile and edge devices. Extensive research focuses on accelerating DNNs while preserving output quality, particularly through model adjustments tailored for resource-limited devices [1-5].
Authors
- Shuochao Yao (ORCID: https://orcid.org/0000-0002-4070-6345)
- Shaohan Hu (ORCID: https://orcid.org/0000-0002-2877-2665)
- Richard Chen (ORCID: https://orcid.org/0000-0002-5912-5620)
- Yuyang Leng (ORCID: https://orcid.org/0009-0008-9376-0880)
- Renyuan Liu (ORCID: https://orcid.org/0000-0001-9710-6116)
- Shilei Tian (ORCID: https://orcid.org/0000-0001-6468-6839)
Institutions
- JPMorgan Chase & Co (United States) (US)
- Advanced Micro Devices (Canada) (CA)
- George Mason University (US)
Publication Details
- Journal
- GetMobile Mobile Computing and Communications
- Published
- 2025-10-29
- DOI
- https://doi.org/10.1145/3774505.3774517
- Primary Topic
- Advanced Neural Network Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00