Robust Learning and Reasoning for Complex Event Forecasting

EVENFLOW will develop hybrid learning techniques for complex event forecasting, which combine deep learning with logic-based learning and reasoning into neurosymbolic forecasting models. The envisioned methods will combine (i) neural representation learning techniques, capable of constructing event-based features from streams of perception-level data with (ii) powerful sMany applications rely on AI to deal with continuously evolving flows of information. The EU-funded EVENFLOW project will develop hybrid learning techniques for forecasting complex changing events. The project will use both neural representation learning techniques that are capable of constructing event-based features from streams of perception-level data as well as symbolic learning and reasoning tools. The online nature of the learning methods will allow them to utilise rich domain knowledge that is becoming available progressively, over time. EVENFLOW will be tested using three cases: oncological forecasting in precision medicine, the safe and efficient behaviour of autonomous transportation robots in smart factories and the reliable life cycle assessment of critical infrastructure.ymbolic learning and reasoning tools, that utilize such features to synthesize high-level, interpretable patterns of critical situations to be forecast.