Catherine Rehwinkel

A use case for the development of a human readable data architecture for parsing film and a sketch for a probabilistic algorithm for teaching it to neural networks.

http://www.catherinerehwinkel.com/itp-nyu/2016/2/18/thesis

Description

The problem: how to preserve of life and love in the face of changing planetary health? A proposed solution begins here: A) A hierarchical data architecture for parsing all measurable aspects of film technique, story, and production at the frame unit-level. B) A sketch for a set of probabilistic algorithms which together work to determine which extracted features in the data architecture are predictors of success in persuading the audience of a director's desired narrative-hypothesis update. C) A a sci-fi screenplay as use case used to develop the algorithms and architecture for parsing causality as reflected in film narrative: An embodied AI learns about human perception of cause and effect by training on these algorithms and then uses her understanding of calculating convincing story events to persuade her human peers to behave in a way which conserves energy across all systems – beginning at the emotional, interpersonal, and societal system levels and ending at the planetary level.

Classes

Thesis