Using HPC and PGAs to Optimize Noisy Computational Models of Cognition

摘要: Cognitive modeling on high performance computing platforms is an emerging field. A preliminary analysis presented the use of parallel processing and genetic algorithms for optimizing fit non-linear, multivariable symbolic models human cognition to experimental data. The effectiveness this optimization methodology illustrated with a prototype model serial arithmetic task built in ACT-R cognitive architecture. results confirm that HPC-based techniques could replace manual used by modelers up until present.