Human-level concept learning through probabilistic program induction

• Published in: Science (American Association for the Advancement of Science) Vol. 350; no. 6266; p. 1332
• Main Authors: Lake, Brenden M, Salakhutdinov, Ruslan, Tenenbaum, Joshua B
• Format: Journal Article
• Language: English
• Published: United States 11.12.2015
• Subjects: Algorithms; Bayes Theorem; Computer Simulation; Concept Formation; Generalization (Psychology); Humans; Machine Learning
• ISSN: 1095-9203
• DOI: 10.1126/science.aab3050

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 outperforming recent deep learning approaches. We also present several "visual Turing tests" probing the model's creative generalization abilities, which in many cases are indistinguishable from human behavior.