701 research outputs found
Glacier : guided locally constrained counterfactual explanations for time series classification
No full textIn machine learning applications, there is a need to obtain predictive models of high performance and, most importantly, to allow end-users and practitioners to understand and act on their predictions. One way to obtain such understanding is via counterfactuals, that provide sample-based explanations in the form of recommendations on which features need to be modified from a test example so that the classification outcome of a given classifier changes from an undesired outcome to a desired one. This paper focuses on the domain of time series classification, more specifically, on defining counterfactual explanations for univariate time series. We propose Glacier, a model-agnostic method for generating locally-constrained counterfactual explanations for time series classification using gradient search either on the original space or on a latent space that is learned through an auto-encoder. An additional flexibility of our method is the inclusion of constraints on the counterfactual generation process that favour applying changes to particular time series points or segments while discouraging changing others. The main purpose of these constraints is to ensure more reliable counterfactuals, while increasing the efficiency of the counterfactual generation process. Two particular types of constraints are considered, i.e., example-specific constraints and global constraints. We conduct extensive experiments on 40 datasets from the UCR archive, comparing different instantiations of Glacier against three competitors. Our findings suggest that Glacier outperforms the three competitors in terms of two common metrics for counterfactuals, i.e., proximity and compactness. Moreover, Glacier obtains comparable counterfactual validity compared to the best of the three competitors. Finally, when comparing the unconstrained variant of Glacier to the constraint-based variants, we conclude that the inclusion of example-specific and global constraints yields a good performance while demonstrating the trade-off between the different metrics. © The Author(s) 2024.This work was funded in part by the Digital Futures cross-disciplinary research centre in Sweden, and the EXTREMUM collaborative project ( https://datascience.dsv.su.se/projects/extremum.html ).</p
Structure Learning with Distributed Parameter Learning for Probabilistic Ontologies
No full textWe consider the problem of learning both the structure and
the parameters of Probabilistic Description Logics under DISPONTE.
DISPONTE (“DIstribution Semantics for Probabilistic ONTologiEs”)
adapts the distribution semantics for Probabilistic Logic Programming
to Description Logics. The system LEAP for “LEArning Probabilistic
description logics” learns both the structure and the parameters of
DISPONTE knowledge bases (KBs) by exploiting the algorithms CELOE
and EDGE. The former stands for “Class Expression Learning for Ontology
Engineering” and it is used to generate good candidate axioms
to add to the KB, while the latter learns the probabilistic parameters
and evaluates the KB. EDGE for “Em over bDds for description loGics
paramEter learning” is an algorithm for learning the parameters of probabilistic
ontologies from data. In order to contain the computational cost,
a distributed version of EDGE called EDGEMR was developed. EDGEMR
exploits the MapReduce (MR) strategy by means of the Message Passing
Interface. In this paper we propose the system LEAPMR. It is a
re-engineered version of LEAP which is able to use distributed parallel
parameter learning algorithms such as EDGEMR
Environmental ethics: values in and duties to the natural world (summarized with commentary by Panagiotis Perros)
No full textSummarized with commentary in Greek by Panagiotis Perros.Environmental ethics stands on a frontier, as radically theoretical as it is applied. Alone, it asks whether there can be nonhuman objects of duty. Animals, plants, endangered species, ecosystems, and even Earth are progressively unfamiliar as objects of duty, and puzzles arise both for theory and practice. Answers to such questions are as urgent as any humans face, and intimately related to the four principal issues on the world agenda: peace, population, development, and environment
CONSTRAINT-BASED MINING OF FREQUENT ARRANGEMENTS OF TEMPORAL INTERVALS
No full textThe problem of discovering frequent arrangements of temporal intervals is studied. It is assumed that the database consists of sequences of events, where an event occurs during a time-interval. The goal is to mine temporal arrangements of event intervals that appear frequently in the database. The motivation of this work is the observation that in practice most events are not instantaneous but occur over a period of time and different events may occur concurrently. Thus, there are many practical applications that require mining such temporal correlations between intervals including the linguistic analysis of annotated data from American Sign Language as well as network and biological data. Two efficient methods to find frequent arrangements of temporal intervals are described; the first one is tree-based and uses depth first search to mine the set of frequent arrangements, whereas the second one is prefix-based. The above methods apply efficient pruning techniques that include a set of constraints consisting of regular expressions and gap constraints that add user-controlled focus into the mining process. Moreover, based on the extracted patterns a standard method for mining association rules is employed that applies different interestingness measures to evaluate the significance of the discovered patterns and rules. The performance of the proposed algorithms is evaluated and compared with other approaches on real (American Sign Language annotations and network data) and large synthetic datasets
R-CAUSTIC: Rippling CAUSTICs underwater Image dataset
No full text<p><strong>Description</strong></p><p>Rippling caustics seem to be the main factor degrading the underwater RGB image quality and affecting the image- based 3D reconstruction process in very shallow waters. These effects are adversely affecting image matching algorithms by throwing off most of them, leading to less accurate matches and causing issues in the Simultaneous Localization and Mapping (SLAM) based navigation of the Remotely Operated Vehicles (ROV) and Autonomous Underwater Vehicles (AUV) on shallow waters. Also, they are the main cause for dissimilarities in the generated textures and orthoimages. In order to fill the gap in the literature regading underwater rippling caustics imagery with real ground truth and reference images, the first real-world underwater caustics benchmark dataset which contains 1465 underwater images is presented. Together with the RGB imagery, the corresponding generated ground truth images are delivered for facilitating the training and testing of machine learning and deep learning methods for image classification. R-CAUSTIC dataset also provides the necessary data to evaluate, at least to some extent, the performance of 3D reconstruction approaches. Data were acquired using a GoPro Hero 4 Black action camera with image dimensions of 4000 x 3000 pixels, focal length of 2.77mm and pixel size of 1.55μm and a tripod. Action cameras are widely used for underwater image acquisition. The dataset was captured in near-shore underwater sites at depths varying from 0.5 to 2m. No artificial light sources were used. Due to the wind, the turbulent surface of the water created dynamic rippling caustics on the seabed. In total 1465 RGB images were collected, separated in 7 different datasets; five of them containing stereo images, one of them tri-stereo images and one consists of multi-stereo imagery acquired in 7 different camera poses.</p><p> </p><p><strong>Publication</strong></p><p>The paper is availbale in Open Access here: https://ieeexplore.ieee.org/document/10172291</p><p><strong>If you use this dataset please cite it as R-CAUSTIC</strong> [Reference].<br>[Reference]: <strong>P. Agrafiotis, K. Karantzalos and A. Georgopoulos, "Seafloor-Invariant Caustics Removal From Underwater Imagery," in </strong><i><strong>IEEE Journal of Oceanic Engineering</strong></i><strong>, vol. 48, no. 4, pp. 1300-1321, Oct. 2023, doi: 10.1109/JOE.2023.3277168.</strong></p><p>BibTeX:</p><p>@ARTICLE{10172291,
author={Agrafiotis, Panagiotis and Karantzalos, Konstantinos and Georgopoulos, Andreas},
journal={IEEE Journal of Oceanic Engineering},
title={Seafloor-Invariant Caustics Removal From Underwater Imagery},
year={2023},
volume={48},
number={4},
pages={1300-1321},
doi={10.1109/JOE.2023.3277168}}</p><p> </p>
R-CAUSTIC: Rippling CAUSTICs underwater Image dataset
No full text<p> </p>
<h3><strong>Version 2 available! Please make sure to download the latest version of the dataset! <br></strong></h3>
<p> </p>
<p><strong>Description</strong></p>
<p>Rippling caustics seem to be the main factor degrading the underwater RGB image quality and affecting the image- based 3D reconstruction process in very shallow waters. These effects are adversely affecting image matching algorithms by throwing off most of them, leading to less accurate matches and causing issues in the Simultaneous Localization and Mapping (SLAM) based navigation of the Remotely Operated Vehicles (ROV) and Autonomous Underwater Vehicles (AUV) on shallow waters. Also, they are the main cause for dissimilarities in the generated textures and orthoimages. In order to fill the gap in the literature regading underwater rippling caustics imagery with real ground truth and reference images, the first real-world underwater caustics benchmark dataset which contains 1465 underwater images is presented. Together with the RGB imagery, the corresponding generated ground truth images are delivered for facilitating the training and testing of machine learning and deep learning methods for image classification. R-CAUSTIC dataset also provides the necessary data to evaluate, at least to some extent, the performance of 3D reconstruction approaches. Data were acquired using a GoPro Hero 4 Black action camera with image dimensions of 4000 x 3000 pixels, focal length of 2.77mm and pixel size of 1.55μm and a tripod. Action cameras are widely used for underwater image acquisition. The dataset was captured in near-shore underwater sites at depths varying from 0.5 to 2m. No artificial light sources were used. Due to the wind, the turbulent surface of the water created dynamic rippling caustics on the seabed. In total 1465 RGB images were collected, separated in 7 different datasets; five of them containing stereo images, one of them tri-stereo images and one consists of multi-stereo imagery acquired in 7 different camera poses.</p>
<p> </p>
<p><strong>Publication</strong></p>
<p>The paper is availbale in Open Access here: https://ieeexplore.ieee.org/document/10172291</p>
<p><strong>If you use this dataset please cite it as R-CAUSTIC</strong> [Reference].<br>[Reference]: <strong>P. Agrafiotis, K. Karantzalos and A. Georgopoulos, "Seafloor-Invariant Caustics Removal From Underwater Imagery," in </strong><em><strong>IEEE Journal of Oceanic Engineering</strong></em><strong>, vol. 48, no. 4, pp. 1300-1321, Oct. 2023, doi: 10.1109/JOE.2023.3277168.</strong></p>
<p>BibTeX:</p>
<p>@ARTICLE{10172291, author={Agrafiotis, Panagiotis and Karantzalos, Konstantinos and Georgopoulos, Andreas}, journal={IEEE Journal of Oceanic Engineering}, title={Seafloor-Invariant Caustics Removal From Underwater Imagery}, year={2023}, volume={48}, number={4}, pages={1300-1321}, doi={10.1109/JOE.2023.3277168}}</p>
<p> </p>
Does genetic diversity on corporate boards lead to improved environmental performance?
No full textElsevier
Journal of International Financial Markets, Institutions and Money
Volume 84, April 2023, 101756
Journal of International Financial Markets, Institutions and Money
Does genetic diversity on corporate boards lead to improved environmental performance?
Author links open overlay panelRenatas Kizys a, Emmanuel C. Mamatzakis b, Panagiotis Tzouvanas c
Show more
Outline
Share
Cite
https://doi.org/10.1016/j.intfin.2023.101756
Get rights and content
Under a Creative Commons license
open access
Highlights
•
We examine the effect of boards’ genetic diversity (GENETICD) on corporate ESG performance.
•
ESG performance and disclosures are higher in more genetically diverse firms.
•
The positive GENETICD effect on ESG performance is driven by the environmental pillar.
•
Corporate carbon performance significantly improves with increases in GENETICD.
We study the effects of boards’ genetic diversity on corporate environmental performance. Using a multidimensional information set for 3690 US firms during the period from 2005 to 2019, and three different measures of genetic diversity, we find that, pursuant to the diversity theory, which posits that diversity improves the quality of management decisions and business ethics, genetic diversity leads to improved environmental performance. We also find that genetic diversity improves carbon and governance performance, and ESG disclosure. Particularly, a one percentage point increase in boards’ genetic diversity will increase the carbon performance, measured by the inverse of the carbon emissions to total assets ratio, and environmental performance by 3.54% and 5.57%, respectively. Our results remain robust to different model specifications, while also controlling for endogeneity. In terms of policy implications, results suggest that the key to tackling climate challenges is to promote boards’ genetic diversity
Enhancing clinical name entity recognition based on hybrid deep learning scheme
No full textThis paper describes a novel machine learning approach based on deeper and wider deep learning model, for better feature learning and latent feature discovery for the clinical name entity recognition task. The performance evaluation of the proposed framework with a benchmark clinical NLP dataset, the clinical CLEF eHealth challenge 2016 dataset, has led to promising performance, when assessed in terms of F-measure, Recall and Precision. The Hybrid CNN model with hyperparameter optimization led to an F-score 89 % for the CLEF eHealth 2016 Challenge task involving synthetic nursing handover dataset.</p
Dataset in support of the Southampton doctoral thesis 'The boatbuilding tradition of the Aegean during the Late Neolithic – Early Bronze Age periods. Typological classification, digital reconstruction and seakeeping assessment'
No full textDataset in support of the Southampton doctoral thesis 'The boatbuilding tradition of the Aegean during the Late Neolithic – Early Bronze Age periods. Typological classification, digital reconstruction and seakeeping assessment' Appendix D - Resistance data and Appendix C - Stability data.
This dataset is focused on two appendices:
Appendix D - Resistance data. D.1 Resistance data produced by the author via MAXSURF Resistance for this thesis.
Appendix C - Stability data
C1. Stability data – STIX and ISO criteria, produced by the author via MAXSURF Stability software for his thesis
This research was funded by Southampton Marine and Maritime Institute (SMMI), Vice-Chancellor's Scholarship, Greek Archaeological Committee UK (GACUK)
</span
- …
