EchoVPR: Echo State Networks for Visual Place Recognition
Authors:
Anil Ozdemir,
Mark Scerri,
Andrew B. Barron,
Andrew Philippides,
Michael Mangan,
Eleni Vasilaki,
Luca Manneschi
Abstract:
Recognising previously visited locations is an important, but unsolved, task in autonomous navigation. Current visual place recognition (VPR) benchmarks typically challenge models to recover the position of a query image (or images) from sequential datasets that include both spatial and temporal components. Recently, Echo State Network (ESN) varieties have proven particularly powerful at solving m…
▽ More
Recognising previously visited locations is an important, but unsolved, task in autonomous navigation. Current visual place recognition (VPR) benchmarks typically challenge models to recover the position of a query image (or images) from sequential datasets that include both spatial and temporal components. Recently, Echo State Network (ESN) varieties have proven particularly powerful at solving machine learning tasks that require spatio-temporal modelling. These networks are simple, yet powerful neural architectures that--exhibiting memory over multiple time-scales and non-linear high-dimensional representations--can discover temporal relations in the data while still maintaining linearity in the learning time. In this paper, we present a series of ESNs and analyse their applicability to the VPR problem. We report that the addition of ESNs to pre-processed convolutional neural networks led to a dramatic boost in performance in comparison to non-recurrent networks in five out of six standard benchmarks (GardensPoint, SPEDTest, ESSEX3IN1, Oxford RobotCar, and Nordland), demonstrating that ESNs are able to capture the temporal structure inherent in VPR problems. Moreover, we show that models that include ESNs can outperform class-leading VPR models which also exploit the sequential dynamics of the data. Finally, our results demonstrate that ESNs improve generalisation abilities, robustness, and accuracy further supporting their suitability to VPR applications.
△ Less
Submitted 10 February, 2022; v1 submitted 11 October, 2021;
originally announced October 2021.
A Pragmatic Approach to Regulating Artificial Intelligence: A Technology Regulator's Perspective
Authors:
Joshua Ellul,
Stephen McCarthy,
Trevor Sammut,
Juanita Brockdorff,
Matthew Scerri,
Gordon J. Pace
Abstract:
Artificial Intelligence (AI) and the regulation thereof is a topic that is increasingly being discussed within various fora. Various proposals have been made in literature for defining regulatory bodies and/or related regulation. In this paper, we present a pragmatic approach for providing a technology assurance regulatory framework. To the best knowledge of the authors this work presents the firs…
▽ More
Artificial Intelligence (AI) and the regulation thereof is a topic that is increasingly being discussed within various fora. Various proposals have been made in literature for defining regulatory bodies and/or related regulation. In this paper, we present a pragmatic approach for providing a technology assurance regulatory framework. To the best knowledge of the authors this work presents the first national AI technology assurance legal and regulatory framework that has been implemented by a national authority empowered through law to do so. In aim of both providing assurances where required and not stifling innovation yet supporting it, herein it is proposed that such regulation should not be mandated for all AI-based systems and that rather it should primarily provide a voluntary framework and only be mandated in sectors and activities where required and as deemed necessary by other authorities for regulated and critical areas.
△ Less
Submitted 15 April, 2021;
originally announced May 2021.
Verifying Web Applications: From Business Level Specifications to Automated Model-Based Testing
Authors:
Christian Colombo,
Mark Micallef,
Mark Scerri
Abstract:
One of reasons preventing a wider uptake of model-based testing in the industry is the difficulty which is encountered by developers when trying to think in terms of properties rather than linear specifications. A disparity has traditionally been perceived between the language spoken by customers who specify the system and the language required to construct models of that system. The dynamic natur…
▽ More
One of reasons preventing a wider uptake of model-based testing in the industry is the difficulty which is encountered by developers when trying to think in terms of properties rather than linear specifications. A disparity has traditionally been perceived between the language spoken by customers who specify the system and the language required to construct models of that system. The dynamic nature of the specifications for commercial systems further aggravates this problem in that models would need to be rechecked after every specification change. In this paper, we propose an approach for converting specifications written in the commonly-used quasi-natural language Gherkin into models for use with a model-based testing tool. We have instantiated this approach using QuickCheck and demonstrate its applicability via a case study on the eHealth system, the national health portal for Maltese residents.
△ Less
Submitted 27 March, 2014;
originally announced March 2014.
