The European Conference on Artificial Intelligence (ECAI 04) was located at the Universidad Politechnica de Valencia. ECAI is huge and incredibly diverse with 181 papers on topics ranging from agents to vision perception, 88 posters and 17 workshops and up to six parallel sessions at any time, so it is pretty much impossible for one person to give an overview.
The ECCAI president Rob Milne opened the conference with a brief address in which he enthused about the “golden age” of AI which is opening up with AI technologies are appearing in technologies that affect people's daily lives for the first time.
The opening keynote was by Christian Freska of Universität Bremen who spoke on “Spatial Perception – an AI Perspective”. He argued that spatial knowledge is key to making sense of physical environments and discussed the importance of qualitative reasoning in this field.
I was at ECAI primarily to attend the 4th Workshop on Computational Models of Natural Argument to give a paper on the ScholOnto approach to modelling argumentation in scholarly articles. The workshop opened with an excellent paper given by Fabio Paglieri on “Argumentation and Data-Oriented Belief Revision: on the Two Sided Nature of Epistemic Change”. This focussed on methods for defeasible reasoning between agents comparing the behaviour of a trustful agent (Mary Magdelene) with a distrustful one (St Thomas). Other interesting articles covered dialogue strategies (Tangmin Yuan) and practical reasoning (Katie Atkinson).