This research develops machine learning techniques to improve fibre-optic access networks, enabling faster, more reliable, and lower-cost internet connections. By allowing transmitters to adapt using receiver feedback, it reduces signal distortion, equipment complexity, and power consumption, helping build resilient communication networks capable of supporting future digital societies.

This research explores how artificial intelligence systems can continue learning without forgetting previously acquired knowledge. Instead of erasing old information, the proposed method compresses knowledge into more efficient representations, allowing AI systems such as self-driving cars to adapt safely to new environments while avoiding dangerous performance failures during learning.