Clare Lyle

This study proposes a new approach to maintain learning in AI systems.

A new method enhances decision-making in reinforcement learning through action-conditional predictions.

A new approach improves learning efficiency in reinforcement learning environments.

Weight clipping enhances model performance in deep learning and reinforcement learning.

Examining the impact of periodic activation functions on learning efficiency and generalization.

A new approach helps neural networks learn from shifting data without forgetting past knowledge.