Theory of Minds: Understanding Behavior in Groups through Inverse Planning

Authors: Michael Shum, Max Kleiman-Weiner, Michael L. Littman, Joshua B. Tenenbaum6163-6170

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our two experiments were carried out in a spatial stag-hunt domain. ... All human data was collected on Amazon Mechanical Turk. In Experiment 1 we compare our algorithm against the inferences people made about the underlying structure i.e., who is cooperating with who. In Experiment 2 we compare our algorithm against people s ability to predict the next action in a scene.
Researcher Affiliation Academia Michael Shum Brain and Cognitive Sciences MIT mshum@mit.edu Max Kleiman-Weiner Brain and Cognitive Sciences MIT maxkw@mit.edu Michael L. Littman Computer Science Brown University mlittman@cs.brown.edu Joshua B. Tenenbaum Brain and Cognitive Sciences MIT jbt@mit.edu
Pseudocode No The paper describes computational procedures using text and mathematical notation, but it does not include a clearly labeled 'Pseudocode' or 'Algorithm' block.
Open Source Code No No statement regarding the availability of open-source code for the described methodology or a link to a code repository was found in the paper.
Open Datasets No The paper describes collecting human data on 'Amazon Mechanical Turk' and using a 'spatial stag-hunt domain' for experiments, but it does not provide concrete access information (e.g., specific link, DOI, repository, or formal citation with authors/year) for either the human data or the simulated environment/data used.
Dataset Splits No The paper describes that 'each participant was tested on all nine scenarios' and 'All human data was collected on Amazon Mechanical Turk', but it does not specify explicit training, validation, or test dataset splits (e.g., percentages or sample counts) for any computational model evaluation.
Hardware Specification No The paper does not provide any specific hardware details such as exact GPU/CPU models, processor types, or memory amounts used for running its experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., library names with version numbers or programming language versions) that would be needed to replicate the experiment.
Experiment Setup Yes In the team inference experiment (experiment 1) participants used a continuous slider to report their judgments so β = 1 was used for model comparison while in the action-prediction experiment (experiment 2) subjects made a discrete choice to report their predictions so a higher β = 5 was used.