# A Simple Algorithm for Graph Reconstruction

arXiv: https://arxiv.org/abs/2112.04549 · HTML: https://mlsanity.com/paper/2112.04549/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Dec 2021.

## Built on by (most cited first, 3 of 3)

- [Tight query complexity bounds for learning graph partitions](https://mlsanity.com/paper/2112.07897/) (2021)
- [Exact Learning of Weighted Graphs Using Composite Queries](https://mlsanity.com/paper/2511.14882/) (2025)
- [Converse and Collision-Based Achievability for Node Localization with Hybrid Distance-Spectral Graph Positional Encodings](https://mlsanity.com/paper/2608.30152/) (2026)

## Abstract (arXiv)

How efficiently can we find an unknown graph using distance queries between its vertices? We assume that the unknown graph is connected, unweighted, and has bounded degree. The goal is to find every edge in the graph. This problem admits a reconstruction algorithm based on multi-phase Voronoi-cell decomposition and using tilde O(n3/2) distance queries. In our work, we analyze a simple reconstruction algorithm. We show that, on random Delta-regular graphs, our algorithm uses tilde O(n) distance queries. As by-products, we can reconstruct those graphs using O(log2 n) queries to an all-distances oracle or tilde O(n) queries to a betweenness oracle, and we bound the metric dimension of those graphs by log2 n. Our reconstruction algorithm has a very simple structure, and is highly parallelizable. On general graphs of bounded degree, our reconstruction algorithm has subquadratic query complexity.