# Black-box Generalization of Machine Teaching

arXiv: https://arxiv.org/abs/2206.15205 · HTML: https://mlsanity.com/paper/2206.15205/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 58 references that are in the map.
- First posted on arXiv 30 Jun 2022.

## Builds on (most cited first, 19 of 19)

- [Deep Bayesian Active Learning with Image Data](https://mlsanity.com/paper/1703.02910/) (2017)
- [BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning](https://mlsanity.com/paper/1906.08158/) (2019)
- [Knockoff Nets: Stealing Functionality of Black-Box Models](https://mlsanity.com/paper/1812.02766/) (2018)
- [Variational Adversarial Active Learning](https://mlsanity.com/paper/1904.00370/) (2019)
- [Semi-Supervised Learning with Ladder Networks](https://mlsanity.com/paper/1507.02672/) (2015)
- [Teacher-Student Curriculum Learning](https://mlsanity.com/paper/1707.00183/) (2017)
- [Consistency-based Semi-supervised Active Learning: Towards Minimizing Labeling Cost](https://mlsanity.com/paper/1910.07153/) (2019)
- [ViewAL: Active Learning with Viewpoint Entropy for Semantic Segmentation](https://mlsanity.com/paper/1911.11789/) (2019)
- [Bayesian Batch Active Learning as Sparse Subset Approximation](https://mlsanity.com/paper/1908.02144/) (2019)
- [An Overview of Machine Teaching](https://mlsanity.com/paper/1801.05927/) (2018)
- [Active Learning for Cost-Sensitive Classification](https://mlsanity.com/paper/1703.01014/) (2017)
- [Adversarial Teacher-Student Learning for Unsupervised Domain Adaptation](https://mlsanity.com/paper/1804.00644/) (2018)
- [Domain Adaptation via Teacher-Student Learning for End-to-End Speech Recognition](https://mlsanity.com/paper/2001.01798/) (2020)
- [Beyond Disagreement-based Agnostic Active Learning](https://mlsanity.com/paper/1407.2657/) (2014)
- [Towards Black-box Iterative Machine Teaching](https://mlsanity.com/paper/1710.07742/) (2017)
- [The Teaching Dimension of Linear Learners](https://mlsanity.com/paper/1512.02181/) (2015)
- [Adaptive Region-Based Active Learning](https://mlsanity.com/paper/2002.07348/) (2020)
- [Active Decision Boundary Annotation with Deep Generative Models](https://mlsanity.com/paper/1703.06971/) (2017)
- [Distribution Matching for Machine Teaching](https://mlsanity.com/paper/2105.13809/) (2021)

## Built on by (most cited first, 1 of 1)

- [Can LLMs Learn by Teaching for Better Reasoning? A Preliminary Study](https://mlsanity.com/paper/2406.14629/) (2024)

## Abstract (arXiv)

Hypothesis-pruning maximizes the hypothesis updates for active learning to find those desired unlabeled data. An inherent assumption is that this learning manner can derive those updates into the optimal hypothesis. However, its convergence may not be guaranteed well if those incremental updates are negative and disordered. In this paper, we introduce a black-box teaching hypothesis hT employing a tighter slack term left(1+FT(widehatht)right)Deltat to replace the typical 2Deltat for pruning. Theoretically, we prove that, under the guidance of this teaching hypothesis, the learner can converge into a tighter generalization error and label complexity bound than those non-educated learners who do not receive any guidance from a teacher:1) the generalization error upper bound can be reduced from R(h^*)+4DeltaT-1 to approximately R(hT)+2DeltaT-1, and 2) the label complexity upper bound can be decreased from 4 θleft(TR(h*)+2O(√T)right) to approximately 2θleft(2TR(hT)+3 O(√T)right). To be strict with our assumption, self-improvement of teaching is firstly proposed when hT loosely approximates h^*. Against learning, we further consider two teaching scenarios: teaching a white-box and black-box learner. Experiments verify this idea and show better generalization performance than the fundamental active learning strategies, such as IWAL, IWAL-D, etc.