CS701 - Theory of Computation Course Page Mcqs Q & A Video Downloads Course Category: Computer Science/Information Technology Course Level: Graduate Credit Hours: 3 Pre-requisites: CS402 MTH202

# Course Synopsis

This is the first pure course in theoretical computer science. It discusses some of the fundamental questions about computation. It starts with an overview of the concepts in theory of automata. Then discusses computability theory in detail. After developing concepts in computability theory the course moves forward to complexity theory. Complexity theory is subdivided into time and space complexity. The course first discusses time complexity and after that space complexity is covered in detail.

# Course Learning Outcomes

At the completion of the course, you should be able to answer the following questions:
• What is computation?
• Is there a universal model of computation?
• Can everything be computed?
• Can we identify problems that are not computable?
• What resources are needed to perform a certain computation?
• Can we identify computationally hard problems?

# Course Contents

Introduction, Set Thoery, Sequences, Tuples, Functions, Relations, Graphs, Turing Machines, Enumerators, Dovetailing, The Church-Turing Thesis, Hilbert's Tenth Problem, Decidable Languages, The Acceptance Problem for DFAs, The Halting Problem, Universal TM, Undicidability of the Halting Problem, Linear Bounded Automata, Computation Histories, Context Free Grammars, Russell's Paradox, Emptiness Problem, Post Correspondence Problem, Computable Functions, Reducibility, Recursion Theorem, Logical Theories, Godel's Theorem, Oracles, Turing Reducibility, A definition of Information, Incompressible Strings, Complexity Theory, Big Oh and Little Oh Notations, Time Complexity, Non-Deterministic Time, The Class P, The Class NP, Polynomial Time Verifiers, Subset Sum Problem, Satisfiability, NP-Completness, 3-Color Problem, The Cook-Levin Theorem, Independent Sets Problem, Clique, Vertex Cover, Hamiltonian Path Problem, The Subset Sum Problem, The Traveling Salesman Problem, Space Complexity, Relationship between Space and Time Complexity, PSPACE-Completeness, TQBF, Prove that TQBF is PSPACE-Complete, FORMULA-GAME, Generalized Geography, LOGSPACE Transducer, Prove the Theorem: NL = co-NL.

 Course Instructor Dr. Sarmad Abbasi Ph.D Computer Science Rutgers University, USA Javascript is disable - Webestools - Vote Service (notation module) Books Introduction to the Theory of Computation by Michael Sipser Solution ( Introduction to the Theory of Computation )