Oreilly – Transfer Learning for Natural Language Processing 2021-8

Transfer Learning for Natural Language Processing. In this course, you will learn how to adapt pre-trained machine learning models to solve specialized natural language processing problems using transfer learning. Training deep natural language processing models from scratch is expensive, time-consuming, and requires huge amounts of data. In this course, DARPA researcher Paul Azur introduces advanced transfer learning techniques that you can use to apply customizable pre-trained models to your natural language processing architectures. You will learn how to use transfer learning to achieve superior language understanding results, even with limited labeled data. Most importantly, you will save training time and computational costs.

Build custom natural language processing models in record time, even with limited datasets, using transfer learning! Transfer learning is a machine learning technique for adapting pre-trained machine learning models to solve specialized problems. This powerful approach has revolutionized natural language processing, helping to improve machine translation, business analytics, and natural language generation.

What you will learn:

• Fine-tuning pre-trained models with new domain data
• Choosing the right model to reduce resource use
• Transfer learning for neural network architectures
• Text generation with pre-trained transformers
• Cross-Language Learning Transfer with BERT
• Fundamentals of Exploring Academic Literature Natural Language Processing

This course is suitable for people who:

• Machine learning engineers and data scientists are experienced in natural language processing.