Ptc Inc.

Joseph John Pizonka

Bruce Katz

Ryan Todd Caplan

A computer-implemented method of automating inductive bias selection includes a computer receiving a plurality of examples, each example providing a plurality of feature-value pairs. The computer constructs an inductive bias dataset which correlates each respective example in the plurality of examples with numerical indications of training quality. The numerical indications of training quality for each respective example are generated by creating a plurality of models, with each model corresponding to a distinct set of inductive biases. The training quality for each respective model is evaluated when applied to the respective example. The computer uses the inductive bias dataset to select a plurality of inductive biases for application to one or more new datasets.

Automated methodology for inductive bias selection and adaptive ensemble choice to optimize predictive power

A computer-implemented method for scoring an example with causal information includes a computer system receiving an indication of a goal and applying a predictive model to the example to yield an output score with respect to the goal. The computer system produces causal score for each feature-value pair in the example, each causal score indicating of a relative extent to which the feature-value pair is responsible for influencing the output score. Then, the computer system presents the output score and the causal score for each feature-value pair in the example.

Growing community of inventors

Phoenixville, PA, United States of America

Joseph John Pizonka