Multi-task learning (MTL) and transfer learning (TL) are techniques to overcome the issue of data scarcity when training state-of-the-art neural networks . However, finding beneficial auxiliary datasets for MTL or TL is a time- and resource-consuming trial-and-error approach . We propose new methods to automatically assess the similarity of sequence tagging datasets to identify beneficial auxiliary data for training neural networks. We provide an efficient, open-source implementation of our new methods . We empirically show that our similarity measures correlate with the change in test score of neural networks that use the auxiliary dataset to increase the main task performance of MTL to increase task performance . We provide a efficient, free-to-use implementation of the open-market implementation

Links: PDF - Abstract

Code :

None

Keywords : task - auxiliary - learning - mtl - networks -

Leave a Reply

Your email address will not be published. Required fields are marked *