Snapcase – Regain Control over Your Predictions with Low-Latency Machine Unlearning
Summary: Snapcase treats recommender models as materialized views and delivers sub‑second user‑level unlearning on a 33M‑purchase, 200k‑user dataset via incremental view maintenance in Differential Dataflow. Unique: a custom top‑k algorithm/data structure for aggregating sparse matrix–matrix multiplies to enable interactive removal of user interactions and their influence. (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 6 of 6 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 522 | Differential dataflow | 2013 | CIDR | 0.00021099241 |
| 1,404 | Responsible Data Management | 2020 | VLDB | 0.00012174977 |
| 3,806 | HedgeCut: Maintaining Randomised Trees for Low-Latency Machine Unlearning | 2021 | SIGMOD | 6.7492837e-05 |
| 4,920 | Shared Arrangements: practical inter-query sharing for streaming dataflows | 2020 | VLDB | 5.8241888e-05 |
| 5,433 | "Amnesia" - A Selection of Machine Learning Models That Can Forget User Data Very Fast | 2020 | CIDR | 5.5051607e-05 |
| 7,489 | DeltaBoost: Gradient Boosting Decision Trees with Efficient Machine Unlearning | 2023 | SIGMOD | 4.7180617e-05 |
Previous
Page 1 / 1
Next