This paper proposes a variational self-attention model (VSAM) that employs variational inference to derive self-attention. We model the self-attention vector as random variables by imposing a probabilistic distribution. The self-attention mechanism summarizes source information as an attention vector by weighted sum, where the weights are a learned probabilistic distribution. Compared with conventional deterministic counterpart, the stochastic units incorporated by VSAM allow multi-modal attention distributions. Furthermore, by marginalizing over the latent variables, VSAM is more robust against overfitting. Experiments on the stance detection task demonstrate the superiority of our method.
@article{arxiv.1812.11559,
title = {Variational Self-attention Model for Sentence Representation},
author = {Qiang Zhang and Shangsong Liang and Emine Yilmaz},
journal= {arXiv preprint arXiv:1812.11559},
year = {2020}
}
Comments
arXiv admin note: text overlap with arXiv:1511.06038 by other authors