Demonstration of Krypton: Optimized CNN Inference for Occlusion-based Deep CNN Explanations
Summary: Krypton optimizes occlusion-based CNN explanations by incremental/approximate inference, cutting runtime up to 35x. Leverages classic query-optimization ideas to enable interactive diagnosis of CNN predictions in radiology and natural images. (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. Allen Ordookhanians (University of California San Diego)
- 2. Xin Li (University of California San Diego)
- 3. Supun Nakandala (University of California San Diego)
- 4. Arun Kumar (University of California San Diego)
BibTeX Citation
@article{ordookhanians_vldb19,
title = {{Demonstration of Krypton: Optimized CNN Inference for Occlusion-based Deep CNN Explanations}},
author = {Ordookhanians, Allen and Li, Xin and Nakandala, Supun and Kumar, Arun},
journal = {PVLDB},
series = {{VLDB} '19},
volume = {12},
number = {12},
pages = {1894--1897},
doi = {10.14778/3352063.3352093},
url = {https://doi.org/10.14778/3352063.3352093},
year = {2019}
}Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,157 | Cerebro: A Data System for Optimized Deep Learning Model Selection | 2020 | VLDB | 0.00011924049 |
| 8,979 | Cerebro: A Layered Data Platform for Scalable Deep Learning | 2021 | CIDR | 5.3399615e-05 |
| 13,375 | Reimagining Deep Learning Systems Through the Lens of Data Systems | 2024 | VLDB | - |
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 |
|---|---|---|---|---|
| 2,987 | Incremental and Approximate Inference for Faster Occlusion-based Deep CNN Explanations | 2019 | SIGMOD | 7.8907997e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 3,573 | VISTA: Optimized System for Declarative Feature Transfer from Deep CNNs at Scale | 2020 | SIGMOD |
| 2 | 10,557 | Database Views as Explanations for Relational Deep Learning | 2026 | VLDB |
| 3 | 11,268 | Optimizing Video Queries with Declarative Clues | 2024 | VLDB |
| 4 | 9,217 | GraphGem: Optimized Scalable System for Graph Convolutional Networks | 2021 | SIGMOD |
| 5 | 11,305 | Demonstration of MaskSearch: Efficiently Querying Image Masks for Machine Learning Workflows | 2024 | VLDB |
| 6 | 9,372 | Intermittent Human-in-the-Loop Model Selection using Cerebro: A Demonstration | 2021 | VLDB |
| 7 | 13,424 | POEM: Pattern-Oriented Explanations of CNN Models | 2022 | VLDB |
| 8 | 3,253 | Jointly Optimizing Preprocessing and Inference for DNN-based Visual Analytics | 2021 | VLDB |
| 9 | 11,327 | Demonstration of VCR: A Tabular Data Slicing Approach to Understanding Object Detection Model Performance | 2024 | VLDB |
| 10 | 2,987 | Incremental and Approximate Inference for Faster Occlusion-based Deep CNN Explanations | 2019 | SIGMOD |