Learning Grounded Causal Models

Noah D. Goodman, Vikash K. Mansinghka, Joshua B. TenenbaumeScholarship 2007

Abstract

We address the problem of learning grounded causal models: systems of concepts that are connected by causal relations and explicitly grounded in perception. We present a Bayesian framework for learning these models—both a causal Bayesian network structure over variables and the consequential region of each variable in perceptual space—from dynamic perceptual evidence. Using a novel experimental paradigm we show that humans are able to learn grounded causal models, and that the Bayesian model accounts well for human performance.
BibTeX
@inproceedings{goodman2007learning,
  author = {Noah D. Goodman and Vikash K. Mansinghka and Joshua B. Tenenbaum},
  title = {{Learning Grounded Causal Models}},
  booktitle = {eScholarship},
  year = {2007},
}

Loading related content...

Figure