Advanced Reasoning with Memory-Augmented Neural Networks

01 November 2020 → Ongoing
Regional and community funding: Special Research Fund
Research disciplines
  • Natural sciences
    • Machine learning and decision making
    • Neural, evolutionary and fuzzy computation
Deep Learning Reasoning with neural networks Memory-Augmented Neural Networks
Project description

Deep neural networks lack fundamental capabilities necessary for learning to reason. We will make fundamental changes to existing neural network models that are capable of variable-binding ; a requirement for reasoning and something that conventional neural network models are incapable of. We envision these fundamental changes will support deep learning methods that can solve more complex and generalizable tasks.