Workshop on Artificial Intelligence Trustworthiness  will be held in Yerevan on April 22. The event will be held in a hybrid format. The organizer will be the AI Center of ISP RAS with the participation of SkolTech and the laboratory of MMO FPMI.

The purpose of the workshop is to provide experts with a platform to share the latest developments in the field of AI reliability, identify problems and opportunities for future work and promote cooperation between researchers and practitioners.

Topics of the works provided for the seminar:

• Methods and principles for the integration of AI in critical products and services in a safe, reliable, and secure way;
• Methods for analyzing datasets for detecting anomalies in markup in order to counter attacks on machine learning;
• ML models with certified robustness;
• Model training methods that provide resistance to adversarial attacks;
• Methods for detecting and countering attacks on AI components in intelligent systems;
• Methods for explaining and improving the interpretability of ML models;
• Research on the resistance to attacks of common models, including the typical architecture of artificial neural networks;
• Techniques for Building Trusted Machine Learning Frameworks and Libraries;
• Engineering of innovative industrial products and services integrating AI;
• Large-scale deployment of industrial systems integrating AI;
• Interaction generating confidence between the user and the AI-based system;
• Ethical and societal implications of intelligent system trustworthiness.

Selected articles will be published in main conference proceedings in the Springer CCIS series indexed by Scopus and Web of Science. Details of the workshop are presented on the website. 

The seminar is collocated with the 11th conference on Artificial Intelligence and Natural Language (AINL 2023), which will be held on 20-21 of April in Yerevan.