
Efficient active learning of sparse halfspaces
We study the problem of efficient PAC active learning of homogeneous lin...
read it

Safety Synthesis Sans Specification
We define the problem of learning a transducer S from a target language ...
read it

Learning Using 1Local Membership Queries
Classic machine learning algorithms learn from labelled examples. For ex...
read it

Point Location and Active Learning: Learning Halfspaces Almost Optimally
Given a finite set X ⊂ℝ^d and a binary linear classifier c: ℝ^d →{0,1}, ...
read it

Condition numberfree query and active learning of linear families
We consider the problem of learning a function from samples with ℓ_2bou...
read it

HS^2: Active Learning over Hypergraphs
We propose a hypergraphbased active learning scheme which we term HS^2,...
read it

Active Learning under Label Shift
Distribution shift poses a challenge for active data collection in the r...
read it
Learning Halfspaces With Membership Queries
Active learning is a subfield of machine learning, in which the learning algorithm is allowed to choose the data from which it learns. In some cases, it has been shown that active learning can yield an exponential gain in the number of samples the algorithm needs to see, in order to reach generalization error ≤ϵ. In this work we study the problem of learning halfspaces with membership queries. In the membership query scenario, we allow the learning algorithm to ask for the label of every sample in the input space. We suggest a new algorithm for this problem, and prove it achieves a near optimal label complexity in some cases. We also show that the algorithm works well in practice, and significantly outperforms uncertainty sampling.
READ FULL TEXT
Comments
There are no comments yet.