Human-level concept learning through probabilistic program induction

People learning new concepts can often generalize successfully from just a single example, yet machine learning algorithms typically require tens or hundreds of examples to perform with similar accuracy. People can also use learned concepts in richer ways than conventional algorithms-for action, imagination, and explanation. We present a computational model that captures these human learning abilities for a large class of simple visual concepts: handwritten characters from the world's alphabets. The model represents concepts as simple programs that best explain observed examples under a Bayesian criterion. On a challenging one-shot classification task, the model achieves human-level performance while…

• CI
  Canadian Institute for Advanced Research
• NS
  Natural Sciences and Engineering Research Council of Canada
• OO
  Office of Naval Research

  Awards: N000141310333, W911NF-08-1-0242, W911NF-13-1-2012
• AR
  Army Research Office