• 注册
• 关注：1 内容：21

• 查看作者
• 打赏作者
• 大版主
Lv. 44
VIP
小可爱

Computability theory studies what problems can or cannot be solved by idealized digital computing machines in principle without regard to the amount of computational resources used. It also studies structures and properties of problems so solvable. A problem is computable in principle if an idealized digital computing machine can solve it using a finite amount of time and memory space, however extravagant the amount may be, for example 1000000010000000000 years and 10000000000000100000000000000000000000 bits of memory space. The halting problem is a well-known example of problem that cannot be solved by algorithm even in principle. Although titled "Computability and Complexity", the topic of this course is complexity theory for the most part. Computational complexity theory is a branch of theoretical computer science that studies the amount of computational resources required to solve computational problems. As has been commented by some, "complexity theory" is a bit of a misnomer since it is not really about how "complex" computational problems are, but the term has been firmly established. "Computational efficiency theory" would be an apter term. Computability theory and complexity theory are treated, respectively, in Part 2 and 3 of Sipser's book.

Time complexity studies the amount of time required to solve problems, and space complexity studies the amount of memory space required to solve problems. These are the two main topics of this course. Toward the end, the course will also treat circuit complexity that studies the amount of logic gates (OR, AND, NOT gates) required to solve problems. In order to build rigorous computational complexity theories, it is necessary to start from a precisely defined computing machine and a complexity measure: a quantitative measure of the amount of resources used such as time, space, logic gates. The predominantly used computing machine for time/space complexities is the Turing machine, which is the one used in this course. Time complexity is then measured by the number of state transitions the Turing machine takes, and space complexity is measured by the number of tape cells the Turing machine uses. In circuit complexity, the computing machine is a circuit of logic gates wired together, and we will study two complexity measures of size complexity and depth complexity.

One of the important topics of computational complexity theory is to classify problems by asymptotic analysis of a complexity measure. Classes of problems thus obtained are called complexity classes. For example, the time complexity class P is defined to be the class of problems that can be solved by deterministic Turing machines in O(nk) steps, i.e., in polynomial numbers of steps. Likewise, the space complexity class PSPACE is defined to be the class of problems that can be solved by deterministic Turing machines using O(nk) tape cells, i.e., in polynomial numbers of tape cells. Equally important is to study relations among complexity classes, such as: Is complexity class C1 contained in another complexity class C2? Are they equal? Do they intersect? These will be the main focus of this course.

This course is much less concerned about practically useful algorithms and their time/space complexity analysis, which are topics of courses in design/analysis of algorithms and data structures (CS 313, CS 323, CS 700).

请登录之后再进行评论

WordPress后台-外观-小工具 进行配置小工具

• 做任务
• 全部清空