We describe a Plug-and-Play controllable language generation framework, Plug-and-Blend, that allows a human user to input multiple control codes (topics). In the context of automated story generation, this allows a human user lose or fine grained control of the topics that will appear in the generated story, and can even allow for overlapping, blended topics. We show that our framework, working with different generation models, controls the generation towards given continuous-weighted control codes while keeping the generated sentences fluent, demonstrating strong blending capability.

Lyrics parody swaps one set of words that accompany a melody with a new set of words, preserving the number of syllables per line and the rhyme scheme. Lyrics parody generation is a challenge for controllable text generation. We show how a specialized sampling procedure, combined with backward text generation with XLNet can produce parody lyrics that reliably meet the syllable and rhyme scheme constraints. We introduce the Weird AI Yankovic system and provide a case study evaluation. We conclude with societal implications of neural lyric parody generation.

Training deep reinforcement learning agents complex behaviors in 3D virtual environments requires significant computational resources. This is especially true in environments with high degrees of aliasing, where many states share nearly identical visual features. Minecraft is an exemplar of such an environment. We hypothesize that interactive machine learning (IML), wherein human teachers play a direct role in training through demonstrations, critique, or action advice, may alleviate agent susceptibility to aliasing. However, interactive machine learning is only practical when the number of human interactions is limited, requiring a balance between human teacher effort and agent performance. We conduct experiments with two reinforcement learning algorithms which enable human teachers to give action advice--Feedback Arbitration, and Newtonian Action Advice--under visual aliasing conditions. To assess potential cognitive load per advice type, we vary the accuracy and frequency of various human action advice techniques. The training efficiency, robustness against infrequent and inaccurate advisor input, and sensitivity to aliasing are examined.

The ability to extract sequences of game events for high-resolution e-sport games has traditionally required access to the game's engine. This serves as a barrier to groups who don't possess this access. It is possible to apply deep learning to derive these logs from gameplay video, but it requires computational power that serves as an additional barrier. These groups would benefit from access to these logs, such as small e-sport tournament organizers who could better visualize gameplay to inform both audience and commentators. In this paper we present a combined solution to reduce the required computational resources and time to apply a convolutional neural network (CNN) to extract events from e-sport gameplay videos. This solution consists of techniques to train a CNN faster and methods to execute predictions more quickly. This expands the types of machines capable of training and running these models, which in turn extends access to extracting game logs with this approach. We evaluate the approaches in the domain of DOTA2, one of the most popular e-sports. Our results demonstrate our approach outperforms standard backpropagation baselines.

Machine learning has been applied to a number of creative, design-oriented tasks. However, it remains unclear how to best empower human users with these machine learning approaches, particularly those users without technical expertise. In this paper we propose a general framework for turn-based interaction between human users and AI agents designed to support human creativity, called {co-creative systems}. The framework can be used to better understand the space of possible designs of co-creative systems and reveal future research directions. We demonstrate how to apply this framework in conjunction with a pair of recent human subject studies, comparing between the four human-AI systems employed in these studies and generating hypotheses towards future studies.

Automated rationale generation is an approach for real-time explanation generation whereby a computational model learns to translate an autonomous agent's internal state and action data representations into natural language. Training on human explanation data can enable agents to learn to generate human-like explanations for their behavior. In this paper, using the context of an agent that plays Frogger, we describe (a) how to collect a corpus of explanations, (b) how to train a neural rationale generator to produce different styles of rationales, and (c) how people perceive these rationales. We conducted two user studies. The first study establishes the plausibility of each type of generated rationale and situates their user perceptions along the dimensions of confidence, humanlike-ness, adequate justification, and understandability. The second study further explores user preferences between the generated rationales with regard to confidence in the autonomous agent, communicating failure and unexpected behavior. Overall, we find alignment between the intended differences in features of the generated rationales and the perceived differences by users. Moreover, context permitting, participants preferred detailed rationales to form a stable mental model of the agent's behavior.

Automated game design has remained a key challenge within the field of Game AI. In this paper, we introduce a method for recombining existing games to create new games through a process called conceptual expansion.Prior automated game design approaches have relied on hand-authored or crowd-sourced knowledge, which limits the scope and applications of such systems. Our approach instead relies on machine learning to learn approximate representations of games. Our approach recombines knowledge from these learned representations to create new games via conceptual expansion.We evaluate this approach by demonstrating the ability for the system to recreate existing games. To the best of our knowledge, this represents the first machine learning-based automated game design system.

The ability of digital storytelling agents to evaluate their output is important for ensuring high-quality human-agent interactions. However, evaluating stories remains an open problem. Past evaluative techniques are either model-specific--- which measure features of the model but do not evaluate the generated stories ---or require direct human feedback, which is resource-intensive. We introduce a number of story features that correlate with human judgments of stories and present algorithms that can measure these features. We find this approach results in a proxy for human-subject studies for researchers evaluating story generation systems.

The ability to extract the sequence of game events for a given player's play-through has traditionally required access to the game's engine or source code. This serves as a barrier to researchers, developers, and hobbyists who might otherwise benefit from these game logs. In this paper we present two approaches to derive game logs from game video via convolutional neural networks and transfer learning. We evaluate the approaches in a Super Mario Bros. clone, Mega Man and Skyrim. Our results demonstrate our approach outperforms random forest and other transfer baselines.

Procedural Level Generation via Machine Learning (PLGML), the study of generating game levels with machine learning, has received a large amount of recent academic attention. For certain measures these approaches have shown success at replicating the quality of existing game levels. However, it is unclear the extent to which they might benefit human designers. In this paper we present a framework for co-creative level design with a PLGML agent.