On the Embeddability of Random Walk Distances
Summary: Proposes a per-node property-aware graph embedding for accurate estimation of random-walk distances (hitting/commute time, PPR); shows standard graph coordinate systems misestimate these metrics. After embedding, node queries run in ~8 μs, enabling orders-of-magnitude speedups with minimal application impact. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Xiaohan Zhao
- 2. Adelbert Chang
- 3. Atish Das Sarma
- 4. Haitao Zheng
- 5. Ben Y. Zhao
Incoming Citations (Sorted by Pagerank)
Showing 4 of 4 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,920 | Fast and Unified Local Search for Random Walk Based K-Nearest-Neighbor Query in Large Graphs | 2014 | SIGMOD | 0.00010090791 |
| 4,020 | TopoX: Topology Refactorization for Efficient Graph Partitioning and Processing | 2019 | VLDB | 6.5237459e-05 |
| 7,394 | Efficient Estimation of Pairwise Effective Resistance | 2023 | SIGMOD | 4.7427524e-05 |
| 10,958 | Efficient Approximation of Kemeny’s Constant for Large Graphs | 2024 | SIGMOD | 4.1945683e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 1 of 1 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 886 | Fast Personalized PageRank on MapReduce | 2011 | SIGMOD | 0.00015597161 |
Previous
Page 1 / 1
Next