Deep Learning, Neural Networks and Natural Language Processing

University of Colorado Boulder - Wednesday May 28, 2014 @ 6:00pm MST

NOTE: For folks unable to attend in person register and we will email you a livestream link 2 hours prior to event.

Location: ATLAS - 1125 18th St Bldg 223, Boulder, CO - Room 100

Map: http://goo.gl/maps/XTJ9v

Agenda:

6:00 - 6:20 Schmooze - Food will be served in Lobby.

6:20 - 6:30 Announcements

6:30 - 7:15 Deep Learning with Neural Networks by William Stanton

7:15 - 8:00 Natural Language Processing with Deep Learning by Nick Dronen

8:00 - 9:00 Network at Old Chicago at 1102 Pearl St. (western end of Pearl Street pedestrian mall, directly facing Boulder Bookstore) -- we have set this to be an hour earlier than last time in the hopes that more can come and support our sponsor, Old Chicago in Boulder, as well as of course enjoy meeting fellow data scientists or aspiring data scientists.

See: http://oldchicago.com/locations/boulder (http://bit.ly/ND8Kp)

Deep Learning with Neural Networks -- Abstract

Artificial Neural Networks are machine-learning methods modeled on the human brain. In recent years, Deep Learning has revolutionized the training and interpretation of neural networks, and many people think Deep Learning is the next frontier in Artificial Intelligence. In this talk, I will give a practical introduction to neural networks and Deep Learning. I will also explain how Deep Learning gives some of the best-ever solutions to problems in computer vision, speech recognition, and natural language processing.

Bio

Will Stanton is on the analytics team at Return Path, the world's leading email data company. Before starting at Return Path, Will studied probability in the Department of Mathematics at CU Boulder. Will loves learning and teaching data science. You can find him on LinkedIn ( http://www.linkedin.com/in/willstanton ) or on his personal website ( http://www.williamgstanton.com/ ).

Natural Language Processing with Deep Learning - Abstract

That supervised deep neural networks are powerful devices for predictive modeling is clear from their often state-of-the-art performance on image and speech data sets. They have also proven to be effective with text data sets, sometimes surprisingly so. In our talk we will present recent applications of neural networks to natural language processing tasks. Our goal is to explain both how these methods work and when one might want to use them.

Bio

Nick Dronen is a research scientist at Pearson in Boulder, where he applies machine learning to the problem of evaluating writing proficiency. He holds a BA in philosophy from Minnesota State University and an MS in computer science from the University of Colorado.