WindTunnel: Towards Differentiable ML Pipelines Beyond a Single Model
Summary: WindTunnel translates heterogeneous ML pipelines—including non-differentiable encoders and gradient-boosting trees—into neural modules. It jointly fine-tunes the entire DAG via backpropagation, improving over sequential, operator-wise training. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Gyeong-In Yu (Seoul National University)
- 2. Saeed Amizadeh (Microsoft)
- 3. Sehoon Kim (University of California Berkeley)
- 4. Artidoro Pagnoni (Carnegie Mellon University)
- 5. Ce Zhang (ETH Zurich)
- 6. Byung-Gon Chun (FriendliAI; Seoul National University)
- 7. Markus Weimer (Microsoft)
- 8. Matteo Interlandi (Microsoft)
BibTeX Citation
@article{yu_vldb22,
title = {{WindTunnel: Towards Differentiable ML Pipelines Beyond a Single Model}},
author = {Yu, Gyeong-In and Amizadeh, Saeed and Kim, Sehoon and Pagnoni, Artidoro and Zhang, Ce and Chun, Byung-Gon and Weimer, Markus and Interlandi, Matteo},
journal = {PVLDB},
series = {{VLDB} '22},
volume = {15},
number = {1},
pages = {11--20},
doi = {10.14778/3485450.3485452},
url = {https://doi.org/10.14778/3485450.3485452},
year = {2022}
}Incoming Citations (Sorted by Pagerank)
Showing 4 of 4 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 5,953 | The Tensor Data Platform: Towards an AI-centric Database System | 2023 | CIDR | 6.0309873e-05 |
| 8,914 | CtxPipe: Context-aware Data Preparation Pipeline Construction for Machine Learning | 2024 | SIGMOD | 5.3483178e-05 |
| 9,881 | The Image Calculator: 10x Faster Image-AI Inference by Replacing JPEG with Self-designing Storage Format | 2024 | SIGMOD | 5.2040783e-05 |
| 11,251 | QCore: Data-Efficient, On-Device Continual Calibration for Quantized Models | 2024 | VLDB | 5.093636e-05 |
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 |
|---|---|---|---|---|
| 835 | Scaling Factorization Machines to Relational Data | 2013 | VLDB | 0.00013721583 |
| 3,874 | Tensors: An abstraction for general data processing | 2021 | VLDB | 7.0561161e-05 |
Previous
Page 1 / 1
Next