L (complexity)

In computational complexity theory, L is the complexity class containing decision problems which can be solved by a deterministic Turing machine using a logarithmic amount of memory space. Intuitively, logarithmic space is enough space to hold a constant number of pointers into the input and a logarithmic number of boolean flags.

A generalization of L is NL, which is the class of languages decidable in logarithmic space on a nondeterministic Turing machine. We then trivially have <math>L \subseteq NL<math>. Also, a decider using O(log n) space cannot use more than 2^{O(log n)}=n^O(1) time, because this is the total number of possible configurations; thus, <math>L \subseteq P<math>, where P is the class of problems solvable in deterministic polynomial time.

Every problem in L is complete under log-space reductions; since this is useless, weaker reductions are defined which allow identification of stronger complete problems in L, but there is no generally accepted definition of L-complete.

Important open problems include whether L = P, and whether L = NL.

The related class of function problems is FL. FL is often used to define logspace reductions.

A breakthrough October 2004 paper by Omer Reingold showed that USTCON, the problem of whether there exists a path between two vertices in a given undirected graph, is in L, establishing that L = SL, since USTCON is SL-complete.

One consequence of this is a simple logical characterization of L: it contains precisely those languages expressible in first order logic with an added commutative transitive closure operator (in graph theoretical terms, this turns every connected component into a clique).

References

• Papadimitriou, Computational Complexity Theory.
• Undirected ST-connectivity in Log-Space. Omer Reingold. Electronic Colloquium on Computational Complexity. No. 94.