Engineering Commons^™

Quantum Classifiers For Video Quality Delivery, Tautvydas Lisas, Ruairí De Fréin

Conference papers

Classical classifiers such as the Support Vector Classifier (SVC) struggle to accurately classify video Quality of Delivery (QoD) time-series due to the challenge in constructing suitable decision boundaries using small amounts of training data. We develop a technique that takes advantage of a quantum-classical hybrid infrastructure called Quantum-Enhanced Codecs (QEC). We evaluate a (1) purely classical, (2) hybrid kernel, and (3) purely quantum classifier for video QoD congestion classification, where congestion is either low, medium or high, using QoD measurements from a real networking test-bed. Findings show that the SVC performs the classification task 4% better in the low congestion …

Experimenting An Edge-Cloud Computing Model On The Gpulab Fed4fire Testbed, Vikas Tomer, Sachin Sharma

Conference papers

There are various open testbeds available for testing algorithms and prototypes, including the Fed4Fire testbeds. This demo paper illustrates how the GPULAB Fed4Fire testbed can be used to test an edge-cloud model that employs an ensemble machine learning algorithm for detecting attacks on the Internet of Things (IoT). We compare experimentation times and other performance metrics of our model based on different characteristics of the testbed, such as GPU model, CPU speed, and memory. Our goal is to demonstrate how an edge-computing model can be run on the GPULab testbed. Results indicate that this use case can be deployed seamlessly …

Underestimation Bias And Underfitting In Machine Learning, Padraig Cunningham, Sarah Jane Delany

Conference papers

. Often, what is termed algorithmic bias in machine learning will be due to historic bias in the training data. But sometimes the bias may be introduced (or at least exacerbated) by the algorithm itself. The ways in which algorithms can actually accentuate bias has not received a lot of attention with researchers focusing directly on methods to eliminate bias - no matter the source. In this paper we report on initial research to understand the factors that contribute to bias in classification algorithms. We believe this is important because underestimation bias is inextricably tied to regularization, i.e. measures to …

Exploration Of Approaches To Arabic Named Entity Recognition, Husamelddin Balla, Sarah Jane Delany

Conference papers

Abstract. The Named Entity Recognition (NER) task has attracted significant attention in Natural Language Processing (NLP) as it can enhance the performance of many NLP applications. In this paper, we compare English NER with Arabic NER in an experimental way to investigate the impact of using different classifiers and sets of features including language-independent and language-specific features. We explore the features and classifiers on five different datasets. We compare deep neural network architectures for NER with more traditional machine learning approaches to NER. We discover that most of the techniques and features used for English NER perform well on Arabic …

Evaluating Load Adjusted Learning Strategies For Client Service Levels Prediction From Cloud-Hosted Video Servers, Ruairí De Fréin, Obinna Izima, Mark Davis

Conference papers

Network managers that succeed in improving the accuracy of client video service level predictions, where the video is deployed in a cloud infrastructure, will have the ability to deliver responsive, SLA-compliant service to their customers. Meeting up-time guarantees, achieving rapid first-call resolution, and minimizing time-to-recovery af- ter video service outages will maintain customer loyalty.

To date, regression-based models have been applied to generate these predictions for client machines using the kernel metrics of a server clus- ter. The effect of time-varying loads on cloud-hosted video servers, which arise due to dynamic user requests have not been leveraged to improve prediction …

Video Quality Prediction Under Time-Varying Loads, Obinna Izima, Ruairí De Fréin, Mark Davis

Conference papers

We are on the cusp of an era where we can responsively and adaptively predict future network performance from network device statistics in the Cloud. To make this happen, regression-based models have been applied to learn mappings between the kernel metrics of a machine in a service cluster and service quality metrics on a client machine. The path ahead requires the ability to adaptively parametrize learning algorithms for arbitrary problems and to increase computation speed. We consider methods to adaptively parametrize regularization penalties, coupled with methods for compensating for the effects of the time-varying loads present in the system, namely …

A Review Of Approaches To Solving The Problem Of Bim Search: Towards Intelligence-Assisted Design, Hamed Khademi, Avril Behan

Conference papers

Due to the growing adoption of BIM and the rising popularity of cloud computing, BIM models are increasingly stored in central cloud repositories or Common Data Environments. Effective management and exploitation of these models creates the requirement for BIM retrieval systems. Thus far, the BIM industry has utilized general-purpose, text-based search techniques that operate on BIM metadata. This paper highlights the need for a domain-specific BIM search engine and reviews various approaches to address the problem of BIM search. Three main approaches were identified as context-, geometry-, and content-based BIM retrieval. For a comprehensive BIM retrieval system, all three approaches …

Investigating The Impact Of Unsupervised Feature-Extraction From Multi-Wavelength Image Data For Photometric Classification Of Stars, Galaxies And Qsos, Annika Lindh

Conference papers

Accurate classification of astronomical objects currently relies on spectroscopic data. Acquiring this data is time-consuming and expensive compared to photometric data. Hence, improving the accuracy of photometric classification could lead to far better coverage and faster classification pipelines. This paper investigates the benefit of using unsupervised feature-extraction from multi-wavelength image data for photometric classification of stars, galaxies and QSOs. An unsupervised Deep Belief Network is used, giving the model a higher level of interpretability thanks to its generative nature and layer-wise training. A Random Forest classifier is used to measure the contribution of the novel features compared to a set …

Activist: A New Framework For Dataset Labelling, Jack O'Neill, Sarah Jane Delany, Brian Mac Namee

Conference papers

Acquiring labels for large datasets can be a costly and time-consuming process. This has motivated the development of the semi-supervised learning problem domain, which makes use of unlabelled data — in conjunction with a small amount of labelled data — to infer the correct labels of a partially labelled dataset. Active Learning is one of the most successful approaches to semi-supervised learning, and has been shown to reduce the cost and time taken to produce a fully labelled dataset. In this paper we present Activist; a free, online, state-of-the-art platform which leverages active learning techniques to improve the efficiency of …