1 code implementation • 17 Jan 2024 • Lei Xun, Jonathon Hare, Geoff V. Merrett
In this thesis, we proposed a combined method, a system was developed for DNN performance trade-off management, combining the runtime trade-off opportunities in both algorithms and hardware to meet dynamically changing application performance targets and hardware constraints in real time.
no code implementations • 17 Jan 2024 • Lei Xun, Mingyu Hu, Hengrui Zhao, Amit Kumar Singh, Jonathon Hare, Geoff V. Merrett
Distributed inference is a popular approach for efficient DNN inference at the edge.
3 code implementations • 20 Sep 2021 • Jia Bi, Jonathon Hare, Geoff V. Merrett
When compared to GhostNet, inference latency on the Jetson Nano is improved by 1. 3x and 2x on the GPU and CPU respectively.
no code implementations • 17 Jul 2021 • Hishan Parry, Lei Xun, Amin Sabet, Jia Bi, Jonathon Hare, Geoff V. Merrett
The new reduced design space results in a BLEU score increase of approximately 1% for sub-optimal models from the original design space, with a wide range for performance scaling between 0. 356s - 1. 526s for the GPU and 2. 9s - 7. 31s for the CPU.
no code implementations • 21 Jun 2021 • Amin Sabet, Jonathon Hare, Bashir Al-Hashimi, Geoff V. Merrett
In this paper, we propose temporal early exits to reduce the computational complexity of per-frame video object detection.
1 code implementation • 8 May 2021 • Lei Xun, Long Tran-Thanh, Bashir M Al-Hashimi, Geoff V. Merrett
Machine learning inference is increasingly being executed locally on mobile and embedded platforms, due to the clear advantages in latency, privacy and connectivity.
no code implementations • 8 May 2021 • Lei Xun, Long Tran-Thanh, Bashir M Al-Hashimi, Geoff V. Merrett
Compared to the existing works, our approach can provide up to 2. 36x (energy) and 2. 73x (time) wider dynamic range with a 2. 4x smaller memory footprint at the same compression rate.
1 code implementation • 8 May 2021 • Wei Lou, Lei Xun, Amin Sabet, Jia Bi, Jonathon Hare, Geoff V. Merrett
However, the training process of such dynamic DNNs can be costly, since platform-aware models of different deployment scenarios must be retrained to become dynamic.