Marta Kwiatkowska

Exploring the need for safety and trust in AI systems.

A new model enhances understanding and adaptability in AI decision-making.

A new approach for players in continuous-state partially-observable games using real-time strategies.

This study focuses on how robots can plan despite making mistakes.

A new method improves the safety verification of neural networks in critical systems.

A new method enhances DNN testing by combining feature selection with uncertainty-based prioritization.

Explore how psychological games shape our choices in everyday life.