Abstract:Motivated by certain applications from physics, biochemistry, economics, and computer science, in which the objects under investigation are not accessible because of various limitations, we propose a trial-and-error model to examine algorithmic issues in such situations. Given a search problem with a hidden input, we are asked to find a valid solution, to find which we can propose candidate solutions (trials), and use observed violations (errors), to prepare future proposals. In accordance with our motivating applications, we consider the fairly broad class of constraint satisfaction problems, and assume that errors are signaled by a verification oracle in the format of the index of a violated constraint (with the content of the constraint still hidden). Our discoveries are summarized as follows. On one hand, despite the seemingly very little information provided by the verification oracle, efficient algorithms do exist for a number of important problems. For the Nash, Core, Stable Matching, and SAT problems, the unknown-input versions are as hard as the corresponding known-input versions, up to a factor of polynomial. We further give almost tight bounds on the latter two problems' trial complexities. On the other hand, there are problems whose complexities are substantially increased in the unknown-input model. In particular, no time-efficient algorithms exist (under standard hardness assumptions) for Graph Isomorphism and Group Isomorphism problems. The tools used to achieve these results include order theory, strong ellipsoid method, and some non-standard reductions. Our model investigates the value of information, and our results demonstrate that the lack of input information can introduce various levels of extra difficulty. The model exhibits intimate connections with (and we hope can also serve as a useful supplement to) certain existing learning and complexity theories.

Further Explanations on "SAT Requires Exhaustive Search"

Hard Examples Requiring Exhaustive Search do Exist.

Evaluating the Claims of "SAT Requires Exhaustive Search"

A survey of SAT solver

Between SAT and UNSAT: The Fundamental Difference in CDCL SAT

An Efficient Resolution Based Algorithm for SAT

Explaining SAT Solving Using Causal Reasoning

Solving SAT by Algorithm Transform of Wu’s Method

EXPLORATION OF HARD TO SOLVE 3-SAT PROBLEMS

A Heuristic Local Search Algorithm for Unsatisfiable Cores Extraction

Improving Local Search for Structured SAT Formulas Via Unit Propagation Based Construct and Cut Initialization (short Paper)

Extracting Minimal Unsatisfiable Subformulas in Satisfiability Modulo Theories.

A Heuristic Complete Algorithm for SAT Problem by Approximation Solution

An Algorithm for Solving Satisfiability Problem Based on the Structural Information of Formulas

Even shorter proofs without new variables

A Resolution-Based Interactive Proof System for UNSAT

On the Complexity of Trial and Error

Finding Unsatisfiable Subformulas With Stochastic Method

Complete Boolean Satisfiability Solving Algorithms Based on Local Search

Improved Algorithm of Unsatisfiability-based Maximum Satisfiability.

An Analog SAT Solver Based on a Deterministic Dynamical System: (invited Paper)