posted on 1992-01-01, 00:00authored byNenad Ivezic, James Henry. Garrett, Carnegie Mellon University.Engineering Design Research Center.
Abstract: "The goal of machine learning for synthesis is the acquisition of the relationships between form, function, and behavior properties that satisfy design requirements. The proposed approach creates a function to estimate the probability of each possible value of each design property being used in a given design context. NETSYN uses a connectionist learning approach to acquire and represent this probability estimation function and exhibits good performance for a posed artificial design problem. The objective of future work is to apply NETSYN to realistic design domains, specifically the domain of computer system design as practiced by M1."