Alsatian: Optimizing Model Search for Deep Transfer Learning
Summary: Alsatian exploits shared model blocks, caching, and search-ordering to accelerate transfer-learning model search. This cache-aware, block-sharing approach reduces repeated inference across thousands of candidate models, yielding up to 14x speedups on CV/NLP benchmarks. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Nils Strassenburg (Hasso Plattner Institute; University of Potsdam)
- 2. Boris Glavic (University of Illinois Chicago)
- 3. Tilmann Rabl (Hasso Plattner Institute; University of Potsdam)
BibTeX Citation
@inproceedings{strassenburg_sigmod25,
title = {{Alsatian: Optimizing Model Search for Deep Transfer Learning}},
author = {Strassenburg, Nils and Glavic, Boris and Rabl, Tilmann},
series = {{SIGMOD} '25},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3725264},
url = {https://dl.acm.org/doi/10.1145/3725264},
year = {2025}
}Incoming Citations (Sorted by Pagerank)
Showing 1 of 1 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,988 | Exploring the Benefits of Just-in-time Model Replacement | 2026 | VLDB | 4.9793485e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 8 of 8 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,355 | Principles of Dataset Versioning: Exploring the Recreation/Storage Tradeoff | 2015 | VLDB | 0.0001092646 |
| 1,568 | HELIX: Holistic Optimization for Accelerating Iterative Machine Learning | 2019 | VLDB | 0.0001021302 |
| 4,334 | LIMA: Fine-grained Lineage Tracing and Reuse in Machine Learning Systems | 2021 | SIGMOD | 6.6569314e-05 |
| 5,553 | Sommelier: Curating DNN Models for the Masses | 2022 | SIGMOD | 6.0858703e-05 |
| 5,685 | SmartLite: A DBMS-based Serving System for DNN Inference in Resource-constrained Environments | 2024 | VLDB | 6.037958e-05 |
| 5,792 | Optimizing Machine Learning Workloads in Collaborative Environments | 2020 | SIGMOD | 5.99349e-05 |
| 7,809 | Nautilus: An Optimized System for Deep Transfer Learning over Evolving Training Datasets | 2022 | SIGMOD | 5.4490255e-05 |
| 8,411 | SHiFT: An Efficient, Flexible Search Engine for Transfer Learning | 2023 | VLDB | 5.336291e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 13,689 | Reimagining Deep Learning Systems Through the Lens of Data Systems | 2024 | VLDB |
| 2 | 6,844 | Serving Deep Learning Models with Deduplication from Relational Databases | 2022 | VLDB |
| 3 | 9,333 | Deep Query Optimization | 2019 | SIGMOD |
| 4 | 3,601 | Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines | 2022 | SIGMOD |
| 5 | 5,286 | An Efficient Transfer Learning Based Configuration Adviser for Database Tuning | 2024 | VLDB |
| 6 | 2,514 | Deep Learning for Blocking in Entity Matching: A Design Space Exploration | 2021 | VLDB |
| 7 | 6,156 | Automatic Data Acquisition for Deep Learning | 2021 | VLDB |
| 8 | 6,559 | DeepBase: Deep Inspection of Neural Networks | 2019 | SIGMOD |
| 9 | 7,809 | Nautilus: An Optimized System for Deep Transfer Learning over Evolving Training Datasets | 2022 | SIGMOD |
| 10 | 8,411 | SHiFT: An Efficient, Flexible Search Engine for Transfer Learning | 2023 | VLDB |