dc.citation.conferencePlace |
AT |
- |
dc.citation.conferencePlace |
Sydney |
- |
dc.citation.endPage |
29 |
- |
dc.citation.startPage |
15 |
- |
dc.citation.title |
IEEE Real-Time and Embedded Technology and Applications Symposium |
- |
dc.contributor.author |
Lee, Seulki |
- |
dc.contributor.author |
Nirjon, Shahriar |
- |
dc.date.accessioned |
2024-01-31T23:07:20Z |
- |
dc.date.available |
2024-01-31T23:07:20Z |
- |
dc.date.created |
2021-08-23 |
- |
dc.date.issued |
2020-04-21 |
- |
dc.description.abstract |
We introduce SubFlow-a dynamic adaptation and execution strategy for a deep neural network (DNN), which enables real-time DNN inference and training. The goal of SubFlow is to complete the execution of a DNN task within a timing constraint that may dynamically change while ensuring comparable performance to executing the full network by executing a subset of the DNN at run-time. To this end, we propose two online algorithms that enable SubFlow: 1) dynamic construction of a sub-network which constructs the best subnetwork of the DNN in terms of size and configuration, and 2) time-bound execution which executes the sub-network within a given time budget either for inference or training. We implement and open-source SubFlow by extending TensorFlow with full compatibility by adding SubFlow operations for convolutional and fully-connected layers of a DNN. We evaluate SubFlow with three popular DNN models (LeNet-5, AlexNet, and KWS), which shows that it provides flexible run-time execution and increases the utility of a DNN under dynamic timing constraints, e.g., lx-6.7x range of dynamic execution speed with average -3% of performance (inference accuracy) difference. We also implement an autonomous robot as an example system that uses SubFlow and demonstrate that its obstacle detection DNN is flexibly executed to meet a range of deadlines that varies depending on its running sped. |
- |
dc.identifier.bibliographicCitation |
IEEE Real-Time and Embedded Technology and Applications Symposium, pp.15 - 29 |
- |
dc.identifier.doi |
10.1109/RTAS48715.2020.00-20 |
- |
dc.identifier.issn |
1545-3421 |
- |
dc.identifier.scopusid |
2-s2.0-85086755425 |
- |
dc.identifier.uri |
https://scholarworks.unist.ac.kr/handle/201301/78553 |
- |
dc.language |
영어 |
- |
dc.publisher |
Institute of Electrical and Electronics Engineers |
- |
dc.title |
SubFlow: A Dynamic Induced-Subgraph Strategy Toward Real-Time DNN Inference and Training |
- |
dc.type |
Conference Paper |
- |
dc.date.conferenceDate |
2020-04-21 |
- |