Contextual and multi-class bandits
- Langford & Zhang, 2007. The epoch-greedy algorithm for contextual multi-armed bandits.
- Kakade, Shalev-Shwartz, & Tewari, 2008. Efficient bandit algorithms for online multiclass prediction.
- Agarwal, Hsu, Kale, Langford, Li, & Schapire, 2014. Taming the monster: a fast and simple algorithm for contextual bandits.
- DudÃk, Hofmann, Schapire, Slivkins, & Zoghi, 2015. Contextual dueling bandits.
- Syrgkanis, Krishnamurthy, & Schapire, 2016. Efficient algorithms for adversarial contextual learning.
- Syrgkanis, Luo, Krishnamurthy, & Schapire, 2016. Improved regret bounds for oracle-based adversarial contextual bandits.
- Beygelzimer, Orabona, & Zhang, 2017. Efficient online bandit multiclass learning witih \(\tilde{O}(\sqrt{T})\) regret.
Active learning
- Freund, Seung, Shamir, & Tishby, 1997. Selective sampling using the Query by Committee algorithm.
- Dasgupta, 2004. Analysis of a greedy active learning strategy.
- Beygelzimer, Dasgupta, & Langford, 2009. Importance weighted active learning.
- Awasthi, Feldman, & Kanade, 2012. Learning using local membership queries.
- Lelkes & Reyzin, 2015. Interactive clustering of linear classes and cryptographic lower bounds.
- Awasthi, Balcan, Haghtalab, & Zhang, 2016. Learning and 1-bit compressed sensing under asymmetric noise.
- Huang, Agarwal, Hsu, Langford, & Schapire, 2016. Efficient and parsimonious agnostic active learning.
- Awasthi, Balcan, & Voevodski, 2017. Local algorithms for interactive clustering.
- Tosh & Dasgupta, 2017. Diameter-based active learning.
Explanations and interpretations