Adaptive Submodularity with Varying Query Sets: An Application to Active Multi-Label Learning
Abstract
Adaptive submodular optimization, where a sequence of items is selected adaptively to optimize a submodular function, has been found to have many applications from sensor placement to active learning. In the current paper, we extend this work to the setting of multiple queries at each time step, where the set of available queries is randomly constrained. A primary contribution of this paper is to prove the first near optimal approximation bound for a greedy policy in this setting. A natural application of this framework is to crowd-sourced active learning problem where the set of available experts and examples might vary randomly. We instantiate the new framework for multi-label learning and evaluate it in multiple benchmark domains with promising results.
Cite
Text
Fern et al. "Adaptive Submodularity with Varying Query Sets: An Application to Active Multi-Label Learning." Proceedings of the 28th International Conference on Algorithmic Learning Theory, 2017.Markdown
[Fern et al. "Adaptive Submodularity with Varying Query Sets: An Application to Active Multi-Label Learning." Proceedings of the 28th International Conference on Algorithmic Learning Theory, 2017.](https://mlanthology.org/alt/2017/fern2017alt-adaptive/)BibTeX
@inproceedings{fern2017alt-adaptive,
title = {{Adaptive Submodularity with Varying Query Sets: An Application to Active Multi-Label Learning}},
author = {Fern, Alan and Goetschalckx, Robby and Hamidi-Haines, Mandana and Tadepalli, Prasad},
booktitle = {Proceedings of the 28th International Conference on Algorithmic Learning Theory},
year = {2017},
pages = {577-592},
volume = {76},
url = {https://mlanthology.org/alt/2017/fern2017alt-adaptive/}
}