Seminar Archive

Most of our funders would like to see non-academic "impact" from the projects they support. In this talk, I'll give a personal perspective on impact, looking at both successes such as the Arria NLG spinout company and the simplenlg open-source package, and failures where expected impact did not materialise, such...

Utilising real-world relational signals (e.g., structured knowledge) and understanding different languages are two (of the many) fundamental goals of Artificial General Intelligence. The corresponding explorations, however, have been relatively separate. In this talk, I will introduce our recent research outputs on bridging this gap:  - We discovered the unnoticed connections between...

Patient-directed digital healthcare delivery is becoming increasingly prominent and has received impetus from the recent COVID pandemic. Patient-directed digital healthcare is viewed positively by policy makers as a means to increase healthcare access, reduce costs and increase efficiency. Patient-directed digital healthcare also appears to be popular with certain patient groups....

Many real-world applications involve optimisation of multiple conflicting objectives simultaneously. Examples include maximizing comfort and minimizing cost while buying or rent a home, minimizing the cost and maximizing the durability of constructed structures, and minimizing the traveling cost and maximum cover of distance.  Such multi-objective optimisation problems require solutions that...

A key problem for rational agents is 'what do do next': which goal the agent should be trying to achieve, and which means it should use to achieve it. In the Belief-Desire-Intention approach to agents, this is termed the 'intention progression problem', and has largely been studied in a single...

To realise accountable AI systems, different types of information from a range of sources need to be recorded throughout the system life cycle. In the RAInS project, we argue that knowledge graphs can support capture and audit of such information; however, the creation of such accountability records must be planned...

This talk will provide an overview and main results of a recently finished 4-year £5M EU-H2020 project (Sep 2017 – Aug 2021) entitled “Core Monitoring Techniques and Experimental Validation and Demonstration (Cortex - https://cortex-h2020.eu/)” that involved 20 partners from 11 countries (EU, Japan and US). Georgios was leading activities in...

In this talk I will describe several related strands of research around the use of argumentation and dialogue for human/machine interaction. These include its use in the context of explanation of program behaviour, as an explanatory mechanism for complex opaque systems such as planners and of computational reasoning process such...

In argumentation theory, we study how to model reasoning through the construction and evaluation of arguments using, so called, argumentation semantics. However, this two steps process (generation and evaluation) have been shown to be limiting due to the large number of arguments generated from real-life knowledge bases. Moreover, although it...

A key challenge in Natural Language Generation (NLG) is evaluating the quality of texts produced by an NLG system.   Robust and reliable evaluation is essential to determining whether a new NLG model or system advances state-of-the-art, understanding weaknesses of current NLG systems, and giving users a good understanding of what...