SIFTER: Space-Efficient Value Iteration for Finite-Horizon MDPs
Summary: SIFTER: value-iteration methods for finite-horizon MDPs that trade horizon-dependent memory for time—one variant uses O(√H) space with no time overhead, another uses O(log H) space with logarithmic time overhead. Shows empirical gains over approximation baselines in space/time/solution quality. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 2 of 2 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 2,156 | SkinnerDB: Regret-Bounded Query Evaluation via Reinforcement Learning | 2018 | VLDB | 9.4170209e-05 |
| 2,219 | SkinnerDB: Regret-Bounded Query Evaluation via Reinforcement Learning | 2019 | SIGMOD | 9.2623533e-05 |
Previous
Page 1 / 1
Next