Multiclass classification with bandit feedback using adaptive regularization

Koby Crammer

doi:10.1007/s10994-012-5321-8

Multiclass classification with bandit feedback using adaptive regularization

2012-10-24

36

OA

AI

DOI：10.1007/s10994-012-5321-8

原文链接

原文求助

摘要

En 中文

We present a new multiclass algorithm in the bandit framework, where after making a prediction, the learning algorithm receives only partial feedback, i.e., a single bit indicating whether the predicted label is correct or not, rather than the true label. Our algorithm is based on the second-order Perceptron, and uses upper-confidence bounds to trade-off exploration and exploitation, instead of random sampling as performed by most current algorithms. We analyze this algorithm in a partial adversarial setting, where instances are chosen adversarially, while the labels are chosen according to a linear probabilistic model which is also chosen adversarially. We show a regret of , which improves over the current best bounds of in the fully adversarial setting. We evaluate our algorithm on nine real-world text classification problems and on four vowel recognition tasks, often obtaining state-of-the-art results, even compared with non-bandit online algorithms, especially when label noise is introduced.

Keyword:

Online learning

Upper confidence bound

Regret

Multiclass classification with bandit feedback using adaptive regularization

摘要

期刊

机构