Learning From Stories: Using Crowdsourced Narratives to Train Virtual Agents

Harrison, Brent; Riedl, Mark O.

doi:10.1609/aiide.v12i1.12876

Cited by 4 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our technique intrinsically motivates the agent to explore states that add nodes and edges that are anticipated by the exemplar story. Related, the learning from stories technique by Harrison et al (2016) uses stories to guide RL agents. However, it requires dozens of exemplar stories and each event is treated as a goal in a modular hierarchical policy; we only require a single story and generate a single unified policy.…”

Section: Background and Related Workmentioning

confidence: 99%

Story Shaping: Teaching Agents Human-Like Behavior with Stories

Peng,

Cui,

Zhou

et al. 2023

AIIDE

View full text Add to dashboard Cite

Reward design for reinforcement learning agents can be difficult in situations where one not only wants the agent to achieve some effect in the world but where one also cares about how that effect is achieved. For example, we might wish for an agent to adhere to a tacit understanding of commonsense, align itself to a preference for how to behave for purposes of safety, or taking on a particular role in an interactive game. Storytelling is a mode for communicating tacit procedural knowledge. We introduce a technique, Story Shaping, in which a reinforcement learning agent infers tacit knowledge from an exemplar story of how to accomplish a task and intrinsically rewards itself for performing actions that make its current environment adhere to that of the inferred story world. Specifically, Story Shaping infers a knowledge graph representation of the world state from observations, and also infers a knowledge graph from the exemplar story. An intrinsic reward is generated based on the similarity between the agent's inferred world state graph and the inferred story world graph. We conducted experiments in text-based games requiring commonsense reasoning and shaping the behaviors of agents as virtual game characters.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Story Shaping: Teaching Agents Human-Like Behavior with Stories

Peng,

Cui,

Zhou

et al. 2023

AIIDE

View full text Add to dashboard Cite

show abstract

“…Li et al (2013) utilized crowdsourced data to generate plot-graph based narrative models. This framework was extended by Harrison and Riedl (2016) to learn reward functions for an RL system that controls virtual agents in the interactive narrative Robbery World.…”

Section: Related Workmentioning

confidence: 99%

“…Lee et al (2014) used dynamic Bayesian networks to model director agent decisions in educational interactive narratives. Crowdsourcing approaches have also been ex-amined to derive plot graphs for generating stories from prior players' input (Li et al 2013;Harrison and Riedl 2016). More recently, reinforcement learning techniques have been used to personalize adaptable event sequences in narrative-centered learning environments (Rowe et al 2014;Wang et al 2016).…”

Section: Introductionmentioning

confidence: 99%

Simulating Player Behavior for Data-Driven Interactive Narrative Personalization

Wang

Rowe

Min

et al. 2021

AIIDE

View full text Add to dashboard Cite

Data-driven approaches to interactive narrative personalization show significant promise for applications in entertainment, training, and education. A common feature of data-driven interactive narrative planning methods is that an enormous amount of training data is required, which is rarely available and expensive to collect from observations of human players. An alternative approach to obtaining data is to generate synthetic data from simulated players. In this paper, we present a long short-term memory (LSTM) neural network framework for simulating players to train data-driven interactive narrative planners. By leveraging a small amount of previously collected human player interaction data, we devise a generative player simulation model. A multi-task neural network architecture is proposed to estimate player actions and experiential outcomes from a single model. Empirical results demonstrate that the bipartite LSTM network produces the better-performing player action prediction models than several baseline techniques, and the multi-task LSTM derives comparable player outcome prediction models within a shorter training time. We also find that synthetic data from the player simulation model contributes to training more effective interactive narrative planners than raw human player data alone.

show abstract

“…Similarly, learning methods can be used to train AI systems for human values. If values cannot easily be enumerated by human programmers, they can be learned (Harrison and Riedl, 2016). Some algorithms are discussed here to make the relationship between human learning methods and machine learning methods more clear.…”

mentioning

confidence: 99%

Human Values from Indian Philosophy to Solve AI Alignment Problem

2024

IJFMR

View full text Add to dashboard Cite

The swift progress of artificial intelligence (AI) has presented society with unparalleled opportunities and challenges. With the growing autonomy of AI systems, the critical concern of ensuring their alignment with human values and ethical principles has come to the forefront. The AI alignment problem refers to the challenge of designing AI systems that act in ways that are beneficial and aligned with human intentions and values. In this paper, we explore the potential contributions of human values from Indian philosophy in solving the AI alignment problem. We conclude that it is possible to establish and tailor a finite set of human values derived from Indian philosophy for the purpose of addressing the enduring challenges that AI systems are expected to tackle in their operational tasks.

show abstract

Learning From Stories: Using Crowdsourced Narratives to Train Virtual Agents

Cited by 4 publications

References 13 publications

Story Shaping: Teaching Agents Human-Like Behavior with Stories

Story Shaping: Teaching Agents Human-Like Behavior with Stories

Simulating Player Behavior for Data-Driven Interactive Narrative Personalization

Human Values from Indian Philosophy to Solve AI Alignment Problem

Contact Info

Product

Resources

About