Abstract | ||
---|---|---|
We study the problem of directly optimizing arbitrary non-differentiable task evaluation metrics such as misclassification rate and recall. Our method, named MetricOpt, operates in a black-box setting where the computational details of the target metric are unknown. We achieve this by learning a differentiable value function, which maps compact task-specific model parameters to metric observations... |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/CVPR46437.2021.00024 | 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) |
Keywords | DocType | ISSN |
Measurement,Computational modeling,Image retrieval,Computer architecture,Object detection,Solids,Pattern recognition | Conference | 1063-6919 |
ISBN | Citations | PageRank |
978-1-6654-4509-2 | 1 | 0.35 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Chen Huang | 1 | 11 | 2.83 |
shuangfei zhai | 2 | 99 | 10.00 |
Pengsheng Guo | 3 | 1 | 0.69 |
Joshua Susskind | 4 | 194 | 9.68 |