The area encompasses fundamental and applied topics in AI and data science, with a focus on machine and deep learning, computer vision, natural language modelling, and speech and audio processing.
Fundamental research aims at enhancing trustworthiness (encompassing fairness and explainability), robustness (handling domain and concept shift, incremental learning, and fault-tolerance), and energy efficiency of AI systems.
Research methods focus primarily on the design, optimization and scalability of learning algorithms and architectures, as well as integration with logical and physical models. Design and optimization methods for energy-efficient machine learning, from edge to cloud scenarios, include hardware/software co-optimization. Applied topics include robotics, analysis of graph, time series, spatio-temporal and tabular data, business intelligence, industry 4.0/5.0 and smart cities.
Computer vision is considered in a broad sense, from fast and well-known algorithms employed in industrial applications, to complex application scenarios with stringent requirements, such as autonomous driving and robotics.
Specific research topics in this area are:
- Trustworthy, safe and explainable AI, fairness and bias in AI, concept-drift in machine learning
- Natural Language Processing, audio and speech processing, multimodal learning
- Theory-guided and physics-informed learning
- Graph, time series and spatio-temporal machine learning
- Big Data algorithms and architectures
- Machine learning on tabular data and relational databases
- Cybersecurity, privacy and network traffic analysis
- Machine learning applications (e.g. predictive maintenance, quantitative analysis of financial markets, anomalies and fault detection)
- Urban data science for smart cities
- Data science for social impact
- Data management and business intelligence to support decision making
- Autonomous learning for intelligent systems in an open-ended manner
- Domain adaptation
- Incremental learning
- First person action recognition
- Computer Vision and AI for robotics
- Automatic speech recognition, speaker and language recognition
- Tiny machine learning (e.g., efficient deep learning for edge and cloud applications)
- Robust and fault-resistant machine learning
- Neuro-symbolic AI
- PE6_7 Artificial intelligence, intelligent systems, multi agent systems
- PE6_8 Computer graphics, computer vision, multimedia, computer games
- PE6_9 Human computer interaction and interface, visualisation and natural language processing
- PE6_10 Web and information systems, database systems, information retrieval and digital libraries, data fusion
- PE6_11 Machine learning, statistical data processing and applications using signal processing (e.g. speech, image, video)