Tuesday, January 31, 2017
10:30 AM - 11:15 AM
|Level: ||Technical - Intermediate|
One of the most promising areas of recent deep learning research is the idea of Artificial Creativity. Game development is a key area where these techniques are being applied. This talk will discuss the background of Artificial Creativity in game design along with its applications today in texture generation, mesh generation, and story generation.
Dave Sullivan is the primary architect of Ersatz and an expert in deep learning and neural networks. Ersatz rose out of his frustration with existing tools for deep learning. He wanted something that he could iterate with quickly but found himself getting bogged down by implementation details and unable to experiment quickly enough. Instead, he imagined a tool that would provide a variety of standard models, accept a variety of standard data formats, and return solutions in a variety of formats. It would all work over the internet and be backed by high end GPU servers. It would focus on deep learning and provide tools specific to deep learning workflows. And so Ersatz was born. Dave has a degree from the University of California at Santa Barbara and is an organizer of the San Francisco Neural Network Aficionados meetup group. See more at: http://www.cognitivecomputingforum.com/speakers.html#sullivan.