Hannu Toivonen

Unsupervised Learning of Cross-Lingual Symbol Embeddings Without Parallel Data

2019 · ScholarWorks@UMassAmherst (University of Massachusetts Amherst)

We present a new method for unsupervised learning of multilingual symbol (e.g. character) embeddings, without any parallel data or prior knowledge about correspondences between languages. It is able to exploit similarities across languages between the …