Sequence Length Independent Norm-Based Generalization Bounds for Transformers
This addresses theoretical generalization guarantees for Transformers, particularly relevant for NLP researchers working with long sequences.
The paper provides norm-based generalization bounds for Transformers that are independent of input sequence length, using covering number bounds for bounded linear transformations to bound Rademacher complexity, and validates these findings empirically on a sparse majority dataset.
This paper provides norm-based generalization bounds for the Transformer architecture that do not depend on the input sequence length. We employ a covering number based approach to prove our bounds. We use three novel covering number bounds for the function class of bounded linear transformations to upper bound the Rademacher complexity of the Transformer. Furthermore, we show this generalization bound applies to the common Transformer training technique of masking and then predicting the masked word. We also run a simulated study on a sparse majority data set that empirically validates our theoretical findings.