METHOD AND APPARATUS FOR EXPLORING AN EXPERIMENTAL SPACE

A hybrid learning system is provided for searching an experimental space. A data mart is configured to acquire, store and manipulate at least historicl experimental data, descriptor data, and concurret experimental data. A search engine is designed t use unsupervised learning techniques to select a et of evaluation points representing a correspondig set of experiments to be run, based on data fromthe data mart. A point evaluation mechanism providd with supervised learning modules which perform pedictive processing based on the evaluation pointsselected by the search engine, and a scoring modul performs a rating operation on outputs of the leaning modules to rate the outputs of the learning mdules from best to worst. The data mart search engne and point evaluation mechanism allow for a repeetitive processing to refine an output of potentia solutions the requirement of continually running ctual physical experiments.