Abstract: 

NLP expertise has become increasingly sought-after in the industry. This is hardly a surprise because of the ubiquity and the growing volume of text data across domains, most likely without the value of text data fully distilled. However, for many machine learning practitioners, the NLP field can be intimidating to navigate and keep up, given the assortment of NLP models and the swiftness of innovation. In actuality, new NLP models are often not entirely new and are usually conceptions to improve previous models’ shortcomings; for example, many recent models are variations of BERT-based architectures.

This talk aims to give an overview walkthrough of the suite of NLP methods grounded in neural-network architectures, including recurrent neural networks (RNNs), transformers, and convolutional neural networks (CNNs). We will connect them by diving into their similarities and differences. You will come away from the talk gaining the overview picture of NLP and grasping the theoretical essence that underpins NLP methods. This talk hopes to empower you with the foundational NLP knowledge and reduce the knowledge barrier for you to jumpstart your NLP projects.

Bio: 

Chengyin Eng is a Senior Data Science Consultant on the Machine Learning Practice team at Databricks. She is experienced in developing end-to-end scalable machine learning solutions for cross-functional clients. She also teaches deep learning and ML in production courses and regularly gives talks at universities and conferences. Prior to Databricks, she worked in the life insurance industry, where she contributed to risk modeling and marketing pipelines. She holds an MS in Computer Science from the University of Massachusetts, Amherst. Her Bachelor's degrees were in Statistics and Environmental Studies from Mount Holyoke College.