Peering inside the 'Black Box': understanding and refining deep neural networks with representational similarity analysis

The work begins with an exploration into interpretability within the clinical domain, developing a perturbative technique to identify diagnostically influential sentences in clinical letters. I then transition to more advanced interpretability techniques, introducing Representational Similarity Analysis and linear probing. A particular focus is placed on layer-wise analysis to observe how salient linguistic signals are encoded throughout the network.

Expanding on this, I use these techniques to measure the fine-grained representation of noun-noun compound thematic relations. This toolkit is then applied to Irish morphosyntax, comparing monolingual and multilingual models to demonstrate the monolingual model's stronger encoding of specific linguistic phenomena. Finally, I introduce Representational Similarity Regularisation, a novel approach to inducing alignment between representations and a target signal.

This method aligns models with cognitive signals elicited by natural language, improving performance on semantic textual similarity tasks.<br><br>This thesis provides novel insights into Transformer models and explores using cognitive signals to build more robust human-aligned neural networks.