The geometry of BERT
This work addresses the need for explainability in Transformer models, offering insights that could lead to architectural improvements, though it is incremental in nature.
The paper tackles the problem of understanding BERT's internal mechanisms by analyzing its attention from a theoretical perspective, including local and global behavior, and demonstrates insights through a case study on SARS-CoV-2 variant classification with very high accuracy.
Transformer neural networks, particularly Bidirectional Encoder Representations from Transformers (BERT), have shown remarkable performance across various tasks such as classification, text summarization, and question answering. However, their internal mechanisms remain mathematically obscure, highlighting the need for greater explainability and interpretability. In this direction, this paper investigates the internal mechanisms of BERT proposing a novel perspective on the attention mechanism of BERT from a theoretical perspective. The analysis encompasses both local and global network behavior. At the local level, the concept of directionality of subspace selection as well as a comprehensive study of the patterns emerging from the self-attention matrix are presented. Additionally, this work explores the semantic content of the information stream through data distribution analysis and global statistical measures including the novel concept of cone index. A case study on the classification of SARS-CoV-2 variants using RNA which resulted in a very high accuracy has been selected in order to observe these concepts in an application. The insights gained from this analysis contribute to a deeper understanding of BERT's classification process, offering potential avenues for future architectural improvements in Transformer models and further analysis in the training process.