All Talk Abstracts

Digital technologies, particularly those related to Artificial Intelligence, are dramatically transforming our lives. Urban Water Cycle, that covers all the services related to the extraction, supply and sanitation of waters in a city, is also being transformed by the development of AI data sources and knowledge management tools. And this transformation encounters the Fourth-Revolution that is taking place in the water sector at the same time.

In the talk, some examples at three levels of complexity (technical, management, socio-political) are presented in the form of references to the work done by the author and collaborators after more than 30 years exploring the possibilities of AI in water. Finally, taking into account that, usually, the state of the art analyses this transformation from predominantly technical and positive perspectives, emphasising the benefits of digitalisation, in the talk some risks of digitalisation in water sector are identified and considered.

End-to-end Deep Learning architectures have been successfully applied to solving challenging visual tasks, often surpassing human performance across various domains. These methods continue to improve due to advancements in their design and increased availability of data. In the recycling industry, accurately predicting the properties of materials that affect recyclability is of particular practical interest. This produces insights that can enhance the efficiency of the recycling process and reduce its environmental impact.

In this talk we will explore the common challenges, opportunities and future outlook of applying deep learning and other machine learning methods in this domain.

Today, there is a general consensus that environmental problems, climate change, loss of species, etc. manifesting themselves in many ways like floods, forest fires, shortage and contamination of water resources, deterioration in food production etc., produce severe and urgent threats that require appropriate interventions locally and globally. Computer Science has developed systems of several kinds that provide support to tackle the problems, for instance remote sensing, geographic information systems, or simulation systems, i.e. to the acquisition, storage and retrieval, and generation of data. They can be supportive, but in a limited way: to become effective, data needs to be interpreted, which requires human expertise, and so does the generation of models for simulation.

Knowledge is needed to understand the nature and origin of environmental problems and to determine appropriate plans to mitigate or prevent their negative impact. And AI pursues the goal of acquiring, representing, and generating knowledge to make it available to intelligent problem solvers like diagnostic, planning, and decision support systems. So, can AI make a contribution to solving the pressing problems? What kind of problems? And how?

In the presentation, we discuss potential contributions of AI (both symbolic AI, with knowledge representation and automated reasoning, and data-driven, training-based techniques) to the acquisition, representation, and exploitation of knowledge and discovery of new insights, thus, producing a better understanding of the problems and improvement of intervention planning. We attempt to identify the major challenges and limitations of these contributions, illustrated by some examples and case studies.

The recent developments in Artificial Intelligence opened the door to improve the management of environmentsl problems through many applications addressing various aspects of environment From waste water treatment to predictive maintenance of aerogeneration or preservation of biodiversity AI contribute in several ways to the SDG. In this talk an overciew of the potential of AI in environmental applications will be reviewed through some real projects.