资源分类
》 资源 》论文》Search Improves Label for Active Learning

Search Improves Label for Active Learning

2020-02-05 | |  39 |   38 |   0

Abstract 

We investigate active learning with access to two distinct oracles: L ABEL (which is standard) and S EARCH (which is not). The S EARCH oracle models the situation where a human searches a database to seed or counterexample an existing solution. S EARCH is stronger than L ABEL while being natural to implement in many situations. We show that an algorithm using both oracles can provide exponentially large problem-dependent improvements over L ABEL alone.

上一篇：Reconstructing Parameters of Spreading Models from Partial Observations

下一篇：Bayesian latent structure discovery from multi-neuron recordings

用户评价
全部评价

热门资源

  • The Variational S...

    Unlike traditional images which do not offer in...

  • Learning to Predi...

    Much of model-based reinforcement learning invo...

  • Stratified Strate...

    In this paper we introduce Stratified Strategy ...

  • A Mathematical Mo...

    Direct democracy, where each voter casts one vo...

  • Rating-Boosted La...

    The performance of a recommendation system reli...