Obtaining sufficient number of high quality annotated data for training large deep learning framework such as BERT, is very difficult in real life scenarios. In this video, a new approach has been explained, which overcomes the limitations for training BERT model using only bunch of labeled data by leveraging Semi Supervised Generative Adversarial Network.
link to the paper - https://www.aclweb.org/anthology/2020...
Link to the Python implementation video - • GANBERT for sentiment analysis (Multi-Clas...
On this page of the site you can watch the video online Multi-Class text classification using only few labeled examples | GAN-BERT Explained with a duration of hours minute second in good quality, which was uploaded by the user Sayantan Adak | PhD Life & Learning 21 June 2021, share the link with friends and acquaintances, this video has already been watched 2,588 times on youtube and it was liked by 43 viewers. Enjoy your viewing!