Abstract:Finding the diameter of a graph in general cannot be done in truly subquadratic assuming the Strong Exponential Time Hypothesis (SETH), even when the underlying graph is unweighted and sparse. When restricting to concrete classes of graphs and assuming SETH, planar graphs and minor-free graphs admit truly subquadratic algorithms, while geometric intersection graphs of unit balls, congruent equilateral triangles, and unit segments do not. Unit-disk graphs are one of the major open cases where the complexity of diameter computation remains unknown. More generally, it is conjectured that a truly-subquadratic time algorithm exists for pseudo-disk graphs where each pair of objects has at most two intersections on the boundary.
In this paper, we show a truly-subquadratic algorithm of running time $\tilde{O}(n^{2-1/18})$, for finding the diameter in a unit-disk graph, whose output differs from the optimal solution by at most 1. This is the first algorithm that provides an additive guarantee in distortion, independent of the size or the diameter of the graph. Our algorithm requires two important technical elements. First, we show that for the intersection graph of pseudo-disks, the graph VC-dimension, either of $k$-hop balls or the distance encoding vectors, is 4. Second, we introduce a clique-based $r$-clustering for geometric intersection graphs, which is an analog of the $r$-division construction for planar graphs. We also showcase the new techniques by establishing new results for distance oracles for unit-disk graphs with subquadratic storage and $O(1)$ query time. The results naturally extend to unit $L_1$- or $L_\infty$-disks and fat pseudo-disks of similar size. Last, if the pseudo-disks additionally have bounded ply, we have a truly-subquadratic algorithm to find the exact diameter.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is the calculation of the diameter of unit - disk graphs. Specifically:
1. **Can the diameter of unit - disk graphs be calculated in truly subquadratic time?** (Problem 1.1)
- Currently, for unit - ball graphs in three or more dimensions, unless the Orthogonal Vector Conjecture is false, it is impossible to calculate the diameter in truly subquadratic time. However, the two - dimensional case is fundamentally different from the case of three or more dimensions. Some problems that are difficult in three or more dimensions become much easier in two dimensions. Therefore, the computational complexity of the diameter of unit - disk graphs (the two - dimensional case) remains an open problem.
2. **Can a +β approximation of the diameter of unit - disk graphs be calculated in truly subquadratic time?** (Problem 1.2)
- The paper proposes an algorithm that can find a +1 approximation of the diameter of unit - disk graphs in $\tilde{O}(n^{2-\frac{1}{18}})$ time, that is, the returned diameter differs from the true diameter by no more than 1. This is the first algorithm to provide an additive error guarantee independent of the size or diameter of the graph.
3. **Can a +β approximation of the diameter of pseudo - disk graphs of similar size be calculated in truly subquadratic time?** (Problem 1.3)
- The paper further generalizes the algorithm to pseudo - disk graphs of similar size and can also find a +1 approximation in $\tilde{O}(n^{2-\frac{1}{18}})$ time.
4. **Can the exact diameter of pseudo - disk graphs of similar size with constant complexity and constant ply be calculated in truly subquadratic time?** (Problem 1.4)
- For pseudo - disk graphs of similar size with constant complexity and constant ply, the paper proposes an algorithm that can calculate the exact diameter in $\tilde{O}(k^{\frac{11}{9}}n^{2-\frac{1}{18}})$ time, where $k$ is the ply of the graph. When $k = O(n^{1/22-\epsilon})$, this running time is truly subquadratic.
### Main Results
- **Theorem 1.5**: For any given unweighted unit - disk graph, a +1 approximation of its diameter can be calculated in $\tilde{O}(n^{2-\frac{1}{18}})$ time.
- **Theorem 1.6**: For any given unweighted pseudo - disk graph of similar size, a +1 approximation of its diameter can be calculated in $\tilde{O}(n^{2-\frac{1}{18}})$ time.
- **Theorem 1.7**: For any given unweighted pseudo - disk graph of similar size, if the ply of the graph is $k$, then the exact diameter can be calculated in $\tilde{O}(k^{\frac{11}{9}}n^{2-\frac{1}{18}})$ time.
- **Theorem 1.8**: For any given unweighted unit - disk graph, a distance query data structure can be constructed, which occupies $\tilde{O}(n^{2-\frac{1}{18}})$ space and has a query time of $O(1)$ per query, and the returned distance differs from the true distance by no more than 1.
- **Theorem 1.9**: For any given unweighted pseudo - disk graph of similar size, a distance query data structure can be constructed, which occupies $\tilde{O}(n^{2-\frac{1}{18}})$ space and has a query time of $O(1)$ per query, and the returned distance differs from the true distance by no more than 1.
### Technical Ideas
- **VC Dimension**: The paper proves that the VC dimension of the ball hypergraph of unit - disk graphs and pseudo - disk graphs is 4, not 3. This result provides a pure topological proof of the topological properties of pseudo - disk graphs.
- **Clique - Based r - Clustering**: Since there is no standard r - division in unit - disk graphs, the paper introduces a new concept - clique - based r - clustering.