CLNov 18, 2020

EasyTransfer -- A Simple and Scalable Deep Transfer Learning Platform for NLP Applications

arXiv:2011.09463v30.1020 citationsHas Code
AI Analysis35

This platform addresses the challenge of building easy-to-use and scalable transfer learning toolkits for developers working on industrial-scale NLP applications.

The EasyTransfer platform was developed to simplify and scale deep transfer learning for NLP applications. It achieves a 4.0x speedup compared to community distributed training methods through built-in data and model parallelism strategies and AI compiler optimization.

The literature has witnessed the success of leveraging Pre-trained Language Models (PLMs) and Transfer Learning (TL) algorithms to a wide range of Natural Language Processing (NLP) applications, yet it is not easy to build an easy-to-use and scalable TL toolkit for this purpose. To bridge this gap, the EasyTransfer platform is designed to develop deep TL algorithms for NLP applications. EasyTransfer is backended with a high-performance and scalable engine for efficient training and inference, and also integrates comprehensive deep TL algorithms, to make the development of industrial-scale TL applications easier. In EasyTransfer, the built-in data and model parallelism strategies, combined with AI compiler optimization, show to be 4.0x faster than the community version of distributed training. EasyTransfer supports various NLP models in the ModelZoo, including mainstream PLMs and multi-modality models. It also features various in-house developed TL algorithms, together with the AppZoo for NLP applications. The toolkit is convenient for users to quickly start model training, evaluation, and online deployment. EasyTransfer is currently deployed at Alibaba to support a variety of business scenarios, including item recommendation, personalized search, conversational question answering, etc. Extensive experiments on real-world datasets and online applications show that EasyTransfer is suitable for online production with cutting-edge performance for various applications. The source code of EasyTransfer is released at Github (https://github.com/alibaba/EasyTransfer).

Code Implementations2 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes