4 citations · 9 across the 12 of their papers we have counts for
12 papers
Dynamic DNNs and Runtime Management for Efficient Inference on Mobile/Embedded Devices
Lei Xun, Jonathon Hare, Geoff V. Merrett
Deep neural network (DNN) inference is increasingly being executed on mobile and embedded platforms due to several key advantages in latency, privacy and always-on availability. Ho…
Fluid Dynamic DNNs for Reliable and Adaptive Distributed Inference on Edge Devices
Lei Xun, Mingyu Hu, Hengrui Zhao +3
Distributed inference is a popular approach for efficient DNN inference at the edge. However, traditional Static and Dynamic DNNs are not distribution-friendly, causing system reli…
Improving the Robustness of Neural Multiplication Units with Reversible Stochasticity
Bhumika Mistry, Katayoun Farrahi, Jonathon Hare
Multilayer Perceptrons struggle to learn certain simple arithmetic tasks. Specialist neural modules for arithmetic can outperform classical architectures with gains in extrapolatio…
Dynamic DNNs Meet Runtime Resource Management on Mobile and Embedded Platforms
Lei Xun, Bashir M. Al-Hashimi, Jonathon Hare +1
Deep neural network (DNN) inference is increasingly being executed on mobile and embedded platforms due to low latency and better privacy. However, efficient deployment on these pl…
Physically Embodied Deep Image Optimisation
Daniela Mihai, Jonathon Hare
Physical sketches are created by learning programs to control a drawing robot. A differentiable rasteriser is used to optimise sets of drawing strokes to match an input image, usin…
Learning Division with Neural Arithmetic Logic Modules
Bhumika Mistry, Katayoun Farrahi, Jonathon Hare
To achieve systematic generalisation, it first makes sense to master simple tasks such as arithmetic. Of the four fundamental arithmetic operations (+,-,,), division is…
Shared Visual Representations of Drawing for Communication: How do different biases affect human interpretability and intent?
Daniela Mihai, Jonathon Hare
We present an investigation into how representational losses can affect the drawings produced by artificial agents playing a communication game. Building upon recent advances, we s…
GhostShiftAddNet: More Features from Energy-Efficient Operations
Jia Bi, Jonathon Hare, Geoff V. Merrett
Deep convolutional neural networks (CNNs) are computationally and memory intensive. In CNNs, intensive multiplication can have resource implications that may challenge the ability…
Dynamic Transformer for Efficient Machine Translation on Embedded Devices
Hishan Parry, Lei Xun, Amin Sabet +3
The Transformer architecture is widely used for machine translation tasks. However, its resource-intensive nature makes it challenging to implement on constrained embedded devices,…
Temporal Early Exits for Efficient Video Object Detection
Amin Sabet, Jonathon Hare, Bashir Al-Hashimi +1
Transferring image-based object detectors to the domain of video remains challenging under resource constraints. Previous efforts utilised optical flow to allow unchanged features…
Dynamic-OFA: Runtime DNN Architecture Switching for Performance Scaling on Heterogeneous Embedded Platforms
Wei Lou, Lei Xun, Amin Sabet +3
Mobile and embedded platforms are increasingly required to efficiently execute computationally demanding DNNs across heterogeneous processing elements. At runtime, the available ha…
A Primer for Neural Arithmetic Logic Modules
Bhumika Mistry, Katayoun Farrahi, Jonathon Hare
Neural Arithmetic Logic Modules have become a growing area of interest, though remain a niche field. These modules are neural networks which aim to achieve systematic generalisatio…