Call for paper: Brain-Like Cognition based Continual Learning and Applications

Brain-Like Cognition based Continual Learning and Applications

The fast development of artificial intelligence (AI), e.g. foundation models has enabled a paradigm shift of many application fields, including cognitive computing, cognitive agents and human-machine interaction. However, traditional AI approaches sometimes struggle with the complexities of human-like learning, adaptation, and generalization. Even the foundation models face some significant limitations: they cannot achieve continual, stable knowledge acquisition and accumulation like the human brain due to catastrophic forgetting when encountering new knowledge.

On the other hand, the human brain possesses remarkable lifelong learning capabilities due to its unique memorial mechanisms, that enable the brain to continually absorb new information while preserving existing knowledge, achieving cumulative knowledge growth. Furthermore, the brain’s hierarchical memory system and memory consolidation processes provide crucial insights for addressing catastrophic forgetting.

This special issue aims to explore novel brain like cognition approaches for continual learning, including theoretical analysis, transforming neuroscientific findings into continual learning computational models, how to leverage foundation model technologies to overcome traditional continual learning limitations, and how to evaluate and validate the effectiveness of these novel approaches.

Key topics

· Brain-like cognition-driven approaches to continual learning.
· Brain memory mechanisms for enhancing continual learning in foundation models.
· Brain-inspired multimodal foundation model fine-tuning, reasoning.
· Personalized solutions for cross-task and cross-distribution continual learning.
· Transparency and interpretability in continual learning models.
· Brain-inspired trustworthy distributed federated learning.
· Brain-inspired collaborative paradigms between foundation models and classical small models. · Brain-inspired generative artificial intelligence.
· Applications of continual learning in real-world domains, including healthcare, robotics, autonomous driving, smart education, etc.
· Brain-like continual learning in embodied intelligence systems, including robotics, sensorimotor control, and interactive environments.