<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">
<div style="margin: 0px;"><b>Start date</b>: 1 January 2018 (or as soon as possible thereafter)</div>
<div style="margin: 0px;"><b>Closing date</b>: 10 December 2017</div>
<div style="margin: 0px;"><b>Duration</b>: 24 months</div>
<div style="margin: 0px;"><b>Keywords</b>: Knowledge graphs, conceptual spaces, vector space embeddings, statistical learning, neural networks</div>
<div style="margin: 0px; min-height: 14px;"><br>
</div>
<div style="margin: 0px;">Applications are invited for two postdoctoral research posts at Cardiff University’s School of Computer Science & Informatics in the context of the ERC funded project FLEXILOG. The overall aims of this project are (i) to learn interpretable
vector space embeddings (or conceptual spaces) from a variety of structured and unstructured information sources, and (ii) to exploit these embeddings for improving statistical and symbolic inference from imperfect data. More information about FLEXILOG can
be found on the project website: <a href="http://www.cs.cf.ac.uk/flexilog/">http://www.cs.cf.ac.uk/flexilog/</a></div>
<div style="margin: 0px; min-height: 14px;"><br>
</div>
<div style="margin: 0px; font-family: Arial;"><span style="letter-spacing: 0px;">Specifically, the aim of these posts will be to contribute to one or more of the following:</span></div>
<div style="margin: 0px; font-family: Arial; min-height: 14px;"><span style="letter-spacing: 0px;"></span><br>
</div>
<div style="margin: 0px; font-family: Arial;"><span style="letter-spacing: 0px;">(i) to develop methods for statistical reasoning from sparse relational data, which exploit vector space representations to impose cognitively inspired forms of regularization
(e.g. the fact that concepts tend to correspond to convex regions). </span></div>
<div style="margin: 0px; font-family: Arial;"><span style="letter-spacing: 0px;">(ii) to develop methods for learning modular and interpretable vector space representations of events, which can be used to predict how events will impact the actors involved (and
the entities related to them), as well as the likelihood of related future events.</span></div>
<div style="margin: 0px; font-family: Arial;"><span style="letter-spacing: 0px;">(iii) to evaluate these methods in applications such as zero-shot learning, textual entailment, reading comprehension, automated knowledge base completion, and entity retrieval.</span></div>
<div style="margin: 0px; min-height: 14px;"><br>
</div>
<div style="margin: 0px;">Successful candidates are expected to have excellent programming skills, as well as a strong background in natural language processing, machine learning, or knowledge representation. </div>
<div style="margin: 0px; min-height: 14px;"><br>
</div>
<div style="margin: 0px;">Cardiff University is a member of the Russell Group of research universities, and was ranked 5th in the UK based on the quality of research in the 2014 Research Evaluation Framework. The university has a successful School of Computer
Science & Informatics with an international reputation for its teaching and research activities. Cardiff is a strong and vibrant capital city with good transportation links and an excellent range of housing available. </div>
<div style="margin: 0px; min-height: 14px;"><br>
</div>
<div style="margin: 0px;"><b>More information:</b></div>
<div style="margin: 0px;">For more details about the positions, please contact Steven Schockaert (<a href="mailto:SchockaertS1@cardiff.ac.uk">SchockaertS1@cardiff.ac.uk</a>). For instructions on how to apply, please go to <a href="http://www.cardiff.ac.uk/jobs">www.cardiff.ac.uk/jobs</a> and
search for job 6522BR. Please note the requirement to evidence all essential criteria in the supporting statement.</div>
</body>
</html>