Natural Language Processing

Teaching computers to understand language is one of the most challenging problems in AI. This season walks through the evolution of natural language processing, from early statistical models to modern transformers. Along the way, it tackles core NLP tasks — translation, sentiment analysis, information extraction — and highlights both the linguistic theory and practical engineering behind machines that read, write, and speak.

Episodes

• Ancient Text Restoration — Thea Sommerschield joins us this week to discuss the development of Pythia - a machine learning model trained to assist in the reconstruction of ancient language text.
• BERT is Shallow — Tim Niven joins us this week to discuss his work exploring the limits of what BERT can do on certain natural language tasks such as adversarial attacks, compositional learning, and systematic learning
• BERT is Magic — Kyle pontificates on how impressed he is with BERT.
• BERT — Kyle provides a non-technical overview of why Bidirectional Encoder Representations from Transformers (BERT) is a powerful tool for natural language processing projects.
• BLEU — Bilingual evaluation understudy (or BLEU) is a metric for evaluating the quality of machine translation using human translation as examples of acceptable quality results. This metric has become a wide