COLGFeb 7, 2024

On a Combinatorial Problem Arising in Machine Teaching

arXiv:2402.04907v25 citationsh-index: 31ICML
AI Analysis

This resolves a theoretical problem in machine teaching, but it is incremental as it confirms an existing conjecture.

The paper proves a conjecture about the worst-case teaching dimension in a machine teaching model, showing it occurs when the consistency matrix contains binary representations of numbers from zero upward, generalizing a theorem on hypercube edge isoperimetry.

We study a model of machine teaching where the teacher mapping is constructed from a size function on both concepts and examples. The main question in machine teaching is the minimum number of examples needed for any concept, the so-called teaching dimension. A recent paper [7] conjectured that the worst case for this model, as a function of the size of the concept class, occurs when the consistency matrix contains the binary representations of numbers from zero and up. In this paper we prove their conjecture. The result can be seen as a generalization of a theorem resolving the edge isoperimetry problem for hypercubes [12], and our proof is based on a lemma of [10].

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes