no code implementations • 22 Apr 2024 • Bailey J. Eccles, Leon Wong, Blesson Varghese
We develop Reconvene, a system for rapidly generating pruned models suited for edge deployments using structured PaI.
no code implementations • 1 Apr 2024 • Dhananjay Saikumar, Blesson Varghese
Pruning methods, such as iterative magnitude-based pruning (IMP) achieve up to a 90% parameter reduction while retaining accuracy comparable to the original model.
no code implementations • 21 Feb 2024 • Dhananjay Saikumar, Blesson Varghese
NeuroFlux segments a CNN into blocks based on GPU memory usage and further attaches an auxiliary network to each layer in these blocks.
1 code implementation • 13 Sep 2023 • Bailey J. Eccles, Philip Rodgers, Peter Kilpatrick, Ivor Spence, Blesson Varghese
Compared to sparse models, the pruned model variants are up to 5. 14x smaller and have a 1. 67x inference latency speedup, with no compromise to sparse model accuracy.
no code implementations • 28 Oct 2022 • JunKyu Lee, Blesson Varghese, Hans Vandierendonck
This paper analyzes the effects of dynamically varying video contents and detection latency on the real-time detection accuracy of a detector and proposes a new run-time accuracy variation model, ROMA, based on the findings from the analysis.
no code implementations • 25 Apr 2022 • Ayesha Abdul Majeed, Peter Kilpatrick, Ivor Spence, Blesson Varghese
This paper will leverage trade-offs in accuracy, end-to-end latency and downtime for selecting the best technique given user-defined objectives (accuracy, latency and downtime thresholds) when an edge node fails.
no code implementations • 27 Nov 2021 • Aaron Yi Ding, Ella Peltonen, Tobias Meuser, Atakan Aral, Christian Becker, Schahram Dustdar, Thomas Hiessl, Dieter Kranzlmuller, Madhusanka Liyanage, Setareh Magshudi, Nitinder Mohan, Joerg Ott, Jan S. Rellermeyer, Stefan Schulte, Henning Schulzrinne, Gurkan Solmaz, Sasu Tarkoma, Blesson Varghese, Lars Wolf
Based on the collective input of Dagstuhl Seminar (21342), this paper presents a comprehensive discussion on AI methods and capabilities in the context of edge computing, referred as Edge AI.
1 code implementation • 2 Nov 2021 • Rehmat Ullah, Di wu, Paul Harvey, Peter Kilpatrick, Ivor Spence, Blesson Varghese
Our empirical results on the CIFAR10 dataset, with both balanced and imbalanced data distribution, support our claims that FedFly can reduce training time by up to 33% when a device moves after 50% of the training is completed, and by up to 45% when 90% of the training is completed when compared to state-of-the-art offloading approach in FL.
1 code implementation • 9 Jul 2021 • Di wu, Rehmat Ullah, Paul Harvey, Peter Kilpatrick, Ivor Spence, Blesson Varghese
Further, FedAdapt adopts reinforcement learning based optimization and clustering to adaptively identify which layers of the DNN should be offloaded for each individual device on to a server to tackle the challenges of computational heterogeneity and changing network bandwidth.
1 code implementation • 5 May 2021 • Hyunho Ahn, Munkyu Lee, Cheol-Ho Hong, Blesson Varghese
For example, applications underpinned by deep neural networks (DNN) models can be sliced and distributed across the IIoT device and the edge of the network for improving the overall performance of inference and for enhancing privacy of the input data, such as industrial product images.
no code implementations • 8 Mar 2021 • Jason Kennedy, Blesson Varghese, Carlos Reaño
This is achieved by offloading computationally intensive workloads, such as deep learning from user devices to the edge.
Edge-computing Distributed, Parallel, and Cluster Computing Networking and Internet Architecture
2 code implementations • 8 Aug 2020 • Luke Lockhart, Paul Harvey, Pierre Imai, Peter Willis, Blesson Varghese
This is because the combination of potential target hardware resources that maximizes performance and the sequence of layers of the DNN that should be distributed across the target resources needs to be determined, while accounting for user-defined objectives/constraints for partitioning.
1 code implementation • 4 Aug 2020 • Francis McNamee, Schahram Dustadar, Peter Kilpatrick, Weisong Shi, Ivor Spence, Blesson Varghese
However, there is limited understanding of: (a) whether and how evolving operational conditions (increased CPU and memory utilization at the edge or reduced data transfer rates between the edge and the cloud) affect the performance of already deployed DNNs, and (b) whether a new partition configuration is required to maximize performance.
1 code implementation • 19 Sep 2018 • Nan Wang, Michail Matthaiou, Dimitrios S. Nikolopoulos, Blesson Varghese
When compared to executing applications on the Edge servers without dynamic vertical scaling, static priorities and dynamic priorities reduce SLO violation rates of requests by up to 4% and 12% for the online game, respectively, and in both cases 6% for the face detection workload.
Distributed, Parallel, and Cluster Computing Systems and Control
no code implementations • 7 Sep 2016 • Blesson Varghese, Nan Wang, Sakil Barbhuiya, Peter Kilpatrick, Dimitrios S. Nikolopoulos
Many cloud-based applications employ a data centre as a central server to process data that is generated by edge devices, such as smartphones, tablets and wearables.
Distributed, Parallel, and Cluster Computing
no code implementations • 8 Aug 2013 • Vu Dung Nguyen, Blesson Varghese, Adam Barker
To this end, a framework for analysing and visualising public sentiment from a Twitter corpus is developed.