The University of Malta’s (LLT) is participating in the project Natural Language for Expainable Artificial Intelligence (NL4XAI) , the largest EU-funded project to date to tackle the challenge of explainable artificial intelligence. NL4XAI is a European Training Network.
The participants of this project are training 11 entrepreneurial and innovative early-stage researchers (ESRs) who will then be facing the challenge of designing and implementing a new generation of self-explanatory AI systems.
The ESRs will receive training in natural language generation and processing, argumentation technology and interactive interfaces for explainable AI (XAI).
Additionally, they will receive training in ethical and legal questions of the field and in transdisciplinary skills. Each ESR will work on their individual research project at one of the network’s host organisations and take part in network-wide training events and meetings, as well as in secondments to other host organisations.
Prof Albert Gatt of the the University of Malta’s Institute of Linguistics and Language Technology is supervising one of these early-stage researchers (ESRs), who will be based in Malta, and co-supervising another two, one at the University of Nancy, France and another at the University of Santiago de Compostela, Spain. Prof Gatt is also the chair of the NL4XAI Training and Development Committee, and will be the main host of one of the consortium meetings.
The XAI models being designed will generate interactive explanations on the basis of natural language and visual tools which are intuitively understandable by non-expert users, validated by humans in specific use cases and accessible to all European citizens.
Main outcomes are to be publicly reported and integrated into a common open source software framework for XAI. In addition, some results are also expected to have commercial potential, warranting protection through licenses or patents.
On 4 September, all project members will meet for the first training event, which will be organised by the network coordinators from CiTIUS (Research Center for Intelligent Technology) at the University of Santiago de Compostela. This is being held simultaneously with the digital European Conference on Artificial Intelligence (ECAI 2020), also organised by the NL4AXAI coordinators.
“With the growth of AI and its increasing presence in our daily lives, explainability has become a very real challenge,” said Prof Gatt. “Most of the time, we find ourselves working with complex neural network architectures, typically trained on vast amounts of data. The problem is that it is hard to understand why systems do what they do in specific circumstances. For example, here at the LLT, our focus in NL4XAI is on Vision and Langauge systems, such as those trained to describe images to a vision-impaired user. It is important to understand why such systems choose the words and expressions that they do. Why, for example, is a woman with a child described as a `mother’? Is the system using evidence gleaned from its training data to come to this conclusion? Does the system’s choice always depend on the visual input? And what other inferences is the system capable of making? How justified are they?”
More information on the project and network members may be found .
H2020 Training Network
This is an H2020 MSCA European Training Network research and innovation programme under the Marie Skłodowska-Curie grant agreement No 860621, funded with 2.8M euros. The network is coordinated by the research team at the Research Centre in Intelligent Technologies of the University of Santiago de Compostela (CiTIUS-USC), headed by Senén Barro.
The consortium is made up of eighteen partners from six different European countries (UK, Malta, The Netherlands, Poland, France and Spain). The partners correspond to two national R&D centres (), ten universities (, , L-Universitá ta’ Malta, , , , , , , ) and six private companies (Indra, Accenture, Orange, Wizenoze, Arria, InfoSupport).
