This research is in the area of computational complexity theory. It will continue to develop the complexity theory of real functions. The main idea is to apply the concepts and techniques of discrete NP- completeness theory to prove lower bound results for important classical theorems in real and complex analysis, as well as important computational problems in numerical analysis. The research will study the notions of generalized Kolmogorov complexity and instance complexity in the context of structural complexity theory, and the relationship between these notions and other important ideas in complexity theory: one-way functions, average-case completeness and pseudorandomness.//