Distribution Matching for Machine Teaching

TL;DR

This paper introduces a distribution matching-based approach for machine teaching that efficiently finds optimal teaching examples without needing to know the student's learning parameters, using a cost-controlled optimization process.

Contribution

It proposes a novel distribution matching strategy for machine teaching that operates without explicit knowledge of student parameters, improving efficiency and applicability.

Findings

The strategy effectively finds teaching sets with limited cost.

The method provides closed-form solutions for training examples.

Theoretical and experimental results validate the approach.

Abstract

Machine teaching is an inverse problem of machine learning that aims at steering the student learner towards its target hypothesis, in which the teacher has already known the student's learning parameters. Previous studies on machine teaching focused on balancing the teaching risk and cost to find those best teaching examples deriving the student model. This optimization solver is in general ineffective when the student learner does not disclose any cue of the learning parameters. To supervise such a teaching scenario, this paper presents a distribution matching-based machine teaching strategy. Specifically, this strategy backwardly and iteratively performs the halving operation on the teaching cost to find a desired teaching set. Technically, our strategy can be expressed as a cost-controlled optimization process that finds the optimal teaching examples without further exploring in the parameter distribution of the student learner. Then, given any a limited teaching cost, the training examples will be closed-form. Theoretical analysis and experiment results demonstrate this strategy.