Content

MCML aims to fill the key gap between traditional machine learning and cognitive computing. The mesoscopic cognitive approach, within the framework of rough set’s knowledge granularity, integrates the knowledge reduction capability of rough sets and the hierarchical characteristics of human cognition, enabling a more comprehensive modeling and expression of knowledge reasoning. This workshop will bring together scholars from the fields of artificial intelligence, cognitive science, and rough set to jointly explore the interpretability of the new generation of AI paradigms and promote the transition from data fitting learning to cognitive-enhanced learning

Keywords and Topics

• Theories and foundations of mesoscopic cognitive machine learning (MCML)
• Rough sets, granular computing, and knowledge reduction for mesoscopic cognition
• Modeling intermediate knowledge structures between data-level and system-level reasoning
• Multi-level and hierarchical knowledge representation aligned with human cognition
• Interpretability and explainability of AI from a mesoscopic perspective
• Integration of rough sets, fuzzy sets, and granular computing into cognitive ML
• Cognitive-enhanced learning frameworks based on mesoscopic reasoning
• Cross-level reasoning mechanisms: linking microscopic data to macroscopic decisions
• Mesoscopic cognitive architectures for hybrid intelligence systems
• Applications of MCML in multimodal AI, natural language processing, computer vision, Medical AI and Autonomous Driving

Organizers

• Ye Wang, Chongqing University of Posts and Telecommunications, China
• Tianwen Qian, East China Normal University, China
• Hong Yu, Chongqing University of Posts and Telecommunications, China