Loading Events

« All Events

  • This event has passed.

Deep Learning & Neural Networks: An Applied Example of Object Detection

July 12 @ 6:30 pm - 9:00 pm

PLEASE REMEMBER, You need to complete a registration and pay on the Chicago Booth site to guarantee a seat (Note: registration will close at noon on the day of the event. Walk-ins will be accepted, space permitting)

https://www.cvent.com/events/an-application-of-deep-learning-to-sports-fitness/registration-5ab7aa7bab894655990b26e4a2b67f65.aspx

We hope you all had a safe and fun Fourth. Please join us for the next Chicago Booth Big Data & Analytics Roundtable.

AN APPLICATION OF DEEP LEARNING TO SPORTS FITNESS
Deep Learning and Neural Networks are increasingly being applied to many problems including object detection. Join us for a unique and engaging example where we apply Deep Learning to Yoga poses.

Thursday, July 12 6:30 PM – 9:00 PM

Presented by:
Ashish Pujari, AVP of Analytics Architecture – CNA Insurance
Emily Coppess, Masters, Analytics – University of Chicago
Jay Ong, Masters, Analytics – University of Chicago
Shahbaz Chaudhry, Masters, Analytics – University of Chicago

Event Details
Developments in deep learning models have greatly expanded the range of problems that approached with neural networks, as well as, the range of problems we can hope to tackle. Neural networks are increasingly being applied to problems related to object detection, natural language processing and generation such as identification of hate speech or adding sound to silent movies.

For this talk, we will discuss an application of deep learning to the area of sports fitness – namely, using deep learning to augment the virtual learning experience for yoga. Deep Learning can be used to develop a system capable of providing instructional feedback. We compare two applications of deep learning to yoga pose classification: a one-step neural network classifier and a two-step model consisting of a pose-extracting neural network feeding into standard classification models like SVM and random forest. We show that by avoiding explicit feature engineering, the one-step model is not only more efficient to build but also performs significantly better than the two-step model for the problem of yoga pose detection.
Cost:
Standard pricing: $15 (register by June 5)
At the door: $20 (subject to availability)

For more information:
https://www.chicagobooth.edu/alumni/events/showEvent?eventId=20834

Details

Date:
July 12
Time:
6:30 pm - 9:00 pm
Event Category:
Website:
https://www.meetup.com/Chicago-Booth-Big-Data-Analytics-Round-Table/events/252602065/

Venue

Gleacher Center
450 N Cityfront Plaza Dr
Chicago, US

Organizer

Chicago Booth Big Data & Analytics Roundtable