For this class of problems, the phase transition behavior is illustrated in Fig. 3. Specifically, this shows the cost to solve the problem with a simple chronological backtrack search. The cost is given in terms of the number of search nodes considered until a solution is found. The minimum cost, for a search that proceeds directly to a solution with no backtrack is . The parameter distinguishing underconstrained from overconstrained problems is the ratio of the number of nogoods m at level 2 given by the constraints to the number of items N.
Figure 3: The solid curves show the classical backtrack search cost for randomly generated problems with a prespecified solution as a function of =m/N for N=10 (gray) and 20 (black) and L=N/2. Here m is the number of nogoods selected at level 2 of the search lattice. The cost is the average number of backtrack steps, starting from the empty set, required to find the first solution to the problem, averaged over 1000 problems. The error bars indicate the standard deviation of this estimate of the average value, and in most cases are smaller than the size of the plotted points. For comparison, the dashed curves show the probability for having a solution in randomly generated problems, ranging from 1 at the left to 0 at the right.
Even for these relatively small problems, a peak in the average search cost is evident. Moreover, this peak is near the transition region where random problems change from mostly soluble to mostly insoluble. A simple, but approximate, theoretical value for the location of the transition is given by the point where the expected number of solutions is equal to one [50, 53]. Applying this to the class of problems considered here is straightforward. Specifically, there are complete sets for the problem, as given by Eq. 1. A particular set s of size L will be good, i.e., a solution, only if none of the nogoods selected for the problem are a subset of s. Hence the probability it will be a solution is given by
because there are sets of size 2 from which to choose the m nogoods specified directly by the constraints. The average number of solutions is then just . If we set and , for large N this becomes
where . The predicted transition point is then given by
which is for the case considered here (i.e., ). This closely matches the location of the peak in the search cost for problems with prespecified solution, as shown in Fig. 3, but is about 20% larger than the location of the step in solubility. Furthermore, the theory predicts there is a regime of polynomial average cost for sufficiently few constraints . This is determined by the condition that the expected number of goods at each level in the lattice is monotonically increasing. Repeating the above argument for smaller levels in the lattice, we find that this condition holds up to
which is for .
While these estimates are only approximate, they do indicate that the class of random soluble problems defined here behaves qualitatively and quantitatively the same with respect to the transition behavior as a variety of other, perhaps more realistic, problem classes. This close correspondence with the theory (derived for the limit of large problems), suggests that we are observing the correct transition behavior even with these relatively small problems. Moreover the above approximate theoretical argument suggests that the average cost of general classical search methods scales exponentially with the size of the problem over the full range of . Thus this provides a good test case for the average behavior of the quantum algorithm. As a final observation, it is important to obtain a sufficient number of samples, especially near the transition region. This is because there is considerable variation in problems near the transition, specifically a highly skewed distribution in the number of solutions. In this region, most problems have few solutions but a few have extremely many: enough in fact to give a substantial contribution to the average number of solutions even though such problems are quite rare.