Trusted Artificial Intelligence
This workshop is supported by the project "Romanian Hub for Artificial Intelligence -HRIA", Smart Growth, Digitization and Financial Instruments Program, MySMIS no. 334906.
Topics:
A) The development of key AI technologies that are safe, reliable, and compliant with EU regulations on AI usage, utilizing supervised, self-supervised, unsupervised, and continual learning techniques for the automatic processing and interpretation of various types of data: images, video, text, voice, unstructured data, and numerical data.
B) The development of explainable AI models and systems that provide interpretability and transparency for decision-making in black-box systems, leading to a better understanding of deep learning models and allowing users to gain insights into the underlying factors and reasoning behind the provided results, as well as ensuring confidence in the model's non-discriminatory nature.
C) The development of technologies for customizing large language models (LLMs) and extending them with safer and more domain- or topic-specific functions, bias control, and explainability assurance in LLMs, while simultaneously reducing discriminatory tendencies and assessing the risks associated with LLM-based systems.
Submission:
The papers will be submitted through the conference submission system specifying the workshop title.
Intelligent and Autonomous Systems
This workshop is supported by the project "Romanian Hub for Artificial Intelligence -HRIA", Smart Growth, Digitization and Financial Instruments Program, MySMIS no. 334906.
Topics:
A) The development, implementation, and evaluation of key technologies for precise, robust, and redundant semantic visual perception in autonomous mobile systems (3D reconstruction, detection, segmentation, tracking, localization, behavior prediction, risk assessment, motion planning) such as mobile robots, autonomous vehicles, and drones.
B) The development of machine learning models and technologies for scene understanding, environmental navigation, human-autonomous system interaction, and semantic-level autonomous control from a multi-task perspective, with minimal human annotations.
C) The development of models and technologies for adaptive, explainable, and interpretable behavior of autonomous robots interacting with human users to perform tasks in diverse environments—e.g., public spaces, hospitals, industrial settings—through human-robot collaboration.
D) The development of intelligent control models in aerial, terrestrial, and underwater robotics, as well as cyber-physical systems, by combining advancements in deep learning with nonlinear control techniques.
E) The development of technologies for generating synthetic datasets, adapting them to specific contexts, and using, evaluating, and validating these data for training models required in intelligent autonomous systems.
Submission:
The papers will be submitted through the conference submission system specifying the workshop title.