Lecture 1 – NLU Course Overview | Stanford CS224U: Natural Language Understanding | Spring 2019

For more information about Stanford’s Artificial Intelligence professional and graduate programs, visit: https://stanford.io/ai

Goals of Natural Language Understanding include:
- Gain insights into human cognition
- Develop artificial intelligence agents as assistants or companions
- Solve a major subproblem of AI

What is understanding?
To understand a statement is to:
- determine the truth (with justification)
- calculate its entailments
- take appropriate action in light of it
- translate it into another language


Professor Christopher Potts & Consulting Assistant Professor Bill MacCartney, Stanford University
http://onlinehub.stanford.edu/

Professor Christopher Potts
Professor of Linguistics and, by courtesy, Computer Science
Director, Stanford Center for the Study of Language and Information
http://web.stanford.edu/~cgpotts/

Consulting Assistant Professor Bill MacCartney
Senior Engineering Manager, Apple
https://nlp.stanford.edu/~wcmac/

To follow along with the course schedule and syllabus, visit: http://web.stanford.edu/class/cs224u/

0:00 Introduction
1:34 Goals of NLU
6:17 Technological and cognitive goals
7:57 What is understanding?
15:01 Philosophical debates
19:29 A question of fact, or a question of usage?
20:41 A brief history of NLU
25:49 How do conversational agents work?
28:04 The promise of conversational agents
29:51 The reality of conversational agents
35:26 Reminiscent of Eliza (1966)
36:39 Conversational search at Google
39:44 Application: sentiment analysis
40:39 Twitter prognostication
41:57 Hathaway vs. Hathaway
42:26 Application: automated trading
43:21 The 2008 United Airlines "bankruptcy"
45:49 The 2013 @AP Twitter hack
46:41 NLU: Traditional organization
47:30 Semantic representations
49:23 Big themes for this class
51:05 Course goals