A uniform problem solving in the cognitive algebra of bounded rational agents

Abstract

The $-calculus cognitive process algebra for problem solving provides the support for automatic problem solving and targets intractable and undecidable problems. Consistent with the ideas of anytime algorithms, $-calculus applies the cost performance measures to converge to optimal solutions with minimal problem solving costs. In the paper, we concentrate on a uniform problem solving and its implementation aspects illustrated on two benchmarks from concurrency and machine learning areas.