Humour-in-the-loop: Improvised Theatre with Interactive Machine Learning Systems

Abstract

Improvisation is a form of live theatre where artists perform real-time, dynamic problem solving to collaboratively generate interesting narratives. The main contribution of this thesis is the development of artificial improvisation: improvised theatre performed by humans alongside intelligent machines. First, I present background underlying the art of improvisation and the scientific fields of interactive machine learning and dialogue generation. Then, I present Pyggy, the first experiment on live stage human-machine improvisation and A.L.Ex., the Artificial Language Experiment which addresses several key technical improvements over Pyggy. Improbotics is then presented which details audience evaluation of Turing test-inspired live improvised performance using A.L.Ex. Two novel contributions to machine-assisted narrative generation are then presented and discussed. The first of these contributions, Shaping the Narrative Arc, is a model incorporating an underlying narrative arc to improve response generation. The second contribution, dAIrector, synthesizes a plot graph with contextual information to generate contextual plot points and serve as director. The thesis concludes by discussing public reflections on live artificial improvisation performances from around the world and interesting future directions to explore. My work presents fundamental advances in human-machine interaction through the lens of improvised theatre which is the ideal test bed for collaboration between humans and intelligent machines.