Novelty Generation Framework for AI Agents in Angry Birds Style Physics Games

Gamage, Chathura; Pinto, Vimukthini; Xue, Cheng; Stephenson, Matthew; Zhang, Peng; Renz, Jochen

doi:10.1109/cog52621.2021.9619160

Cited by 8 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Likewise, there may be observations and states that never occur pre-novelty that start to occur post-novelty. This is consistent with a robotics perspective on MDPs where actions and observations are governed by an underlying physics of the real world, even though we experiment within grid worlds and games (Kejriwal and Thomas 2021; Gamage et al 2021).…”

Section: An Ontology Of Novelty For Sequential Decision Makingsupporting

confidence: 85%

“…The Novgridworld environment implements a Minecraftinspired grid world to study novelties in the context of agent-driven object-object interaction (Goel et al 2021). The GNOME Monopoly environment examines multi-agent gameplay in a long term multi-faceted strategic context, while Science Birds has a fewer number of timesteps and in each episode by examining both observational and world novelties (Kejriwal and Thomas 2021;Gamage et al 2021). In both cases, an implementation is provided for outside users.…”

Section: Novelty Background and Related Workmentioning

confidence: 99%

See 1 more Smart Citation

NovGrid: A Flexible Grid World for Evaluating Agent Response to Novelty

Balloch¹,

Lin²,

Hussain³

et al. 2022

Preprint

View full text Add to dashboard Cite

A robust body of reinforcement learning techniques have been developed to solve complex sequential decision making problems. However, these methods assume that train and evaluation tasks come from similarly or identically distributed environments. This assumption does not hold in real life where small novel changes to the environment can make a previously learned policy fail or introduce simpler solutions that might never be found. To that end we explore the concept of novelty, defined in this work as the sudden change to the mechanics or properties of environment. We provide an ontology of for novelties most relevant to sequential decision making, which distinguishes between novelties that affect objects versus actions, unary properties versus non-unary relations, and the distribution of solutions to a task. We introduce NOVGRID, a novelty generation framework built on Mini-Grid, acting as a toolkit for rapidly developing and evaluating novelty-adaptation-enabled reinforcement learning techniques. Along with the core NOVGRIDwe provide exemplar novelties aligned with our ontology and instantiate them as novelty templates that can be applied to many MiniGridcompliant environments. Finally, we present a set of metrics built into our framework for the evaluation of noveltyadaptation-enabled machine-learning techniques, and show characteristics of a baseline RL model using these metrics.

show abstract

Section: An Ontology Of Novelty For Sequential Decision Makingsupporting

confidence: 85%

Section: Novelty Background and Related Workmentioning

confidence: 99%

NovGrid: A Flexible Grid World for Evaluating Agent Response to Novelty

Balloch¹,

Lin²,

Hussain³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Numerous research studies have explored various methods for creating game levels in Angry Birds. Prior investigations have tackled this problem from diverse angles, proposing solutions to intriguing challenges, such as generating stable structures within the physical environment , dynamically adjusting the game levels' difficulty (Stephenson and Renz 2019), creating block structures based on hand-drawn sketches (Stephenson et al 2021), generating deceptive levels to deceive AI agents (Gamage et al 2021a), generating novel scenarios for physics environments (Gamage et al 2023(Gamage et al , 2021b and, most recently, using prompt engineering to create prompts for level generation (Taveekitworachai et al 2023). None of the previous studies has treated the level generation for Angry Birds as a task generation problem that incorporates the physical interactions between the game objects.…”

Section: Physics Based Puzzle Gamesmentioning

confidence: 99%

Physics-Based Task Generation through Causal Sequence of Physical Interactions

Gamage,

Pinto,

Stephenson

et al. 2023

AIIDE

View full text Add to dashboard Cite

Performing tasks in a physical environment is a crucial yet challenging problem for AI systems operating in the real world. Physics simulation-based tasks are often employed to facilitate research that addresses this challenge. In this paper, first, we present a systematic approach for defining a physical scenario using a causal sequence of physical interactions between objects. Then, we propose a methodology for generating tasks in a physics-simulating environment using these defined scenarios as inputs. Our approach enables a better understanding of the granular mechanics required for solving physics-based tasks, thereby facilitating accurate evaluation of AI systems' physical reasoning capabilities. We demonstrate our proposed task generation methodology using the physics-based puzzle game Angry Birds and evaluate the generated tasks using a range of metrics, including physical stability, solvability using intended physical interactions, and accidental solvability using unintended solutions. We believe that the tasks generated using our proposed methodology can facilitate a nuanced evaluation of physical reasoning agents, thus paving the way for the development of agents for more sophisticated real-world applications.

show abstract

“…The success of recent AIs in games such as Go, Chess, Minecraft, and Monopoly ( Silver et al, 2017 ; Goecks et al, 2021b ; Haliem et al, 2021 ) are based on games with complete knowledge of the existing state of the world (i.e., “open” games), whereas wargaming typically includes incomplete (e.g., Starcraft), uncertain, and/or deceptive information about the operational environment ( Vinyals et al, 2019 ). Uncertainty can also arise from changing physics or other environmental rules, as has been explored with Angry Birds ( Gamage et al, 2021 ). The lack of knowledge makes it difficult for AI agents to calculate the risk-reward profiles of future actions, due to the uncertainty in the state of the world, state of the different actors, and the effects of the actions taken ( Cassenti and Kaplan, 2021 ).…”

Section: Required Advancements For Future Military Decision Making Pr...mentioning

confidence: 99%

The Future of Collaborative Human-Artificial Intelligence Decision-Making for Mission Planning

et al. 2022

View full text Add to dashboard Cite

In an increasingly complex military operating environment, next generation wargaming platforms can reduce risk, decrease operating costs, and improve overall outcomes. Novel Artificial Intelligence (AI) enabled wargaming approaches, based on software platforms with multimodal interaction and visualization capacity, are essential to provide the decision-making flexibility and adaptability required to meet current and emerging realities of warfighting. We highlight three areas of development for future warfighter-machine interfaces: AI-directed decisional guidance, computationally informed decision-making, and realistic representations of decision spaces. Progress in these areas will enable development of effective human-AI collaborative decision-making, to meet the increasing scale and complexity of today’s battlespace.

show abstract

Novelty Generation Framework for AI Agents in Angry Birds Style Physics Games

Cited by 8 publications

References 14 publications

NovGrid: A Flexible Grid World for Evaluating Agent Response to Novelty

NovGrid: A Flexible Grid World for Evaluating Agent Response to Novelty

Physics-Based Task Generation through Causal Sequence of Physical Interactions

The Future of Collaborative Human-Artificial Intelligence Decision-Making for Mission Planning

Contact Info

Product

Resources

About