Iterative Teacher-Aware Learning

TL;DR

This paper introduces an iterative teacher-aware learning algorithm that incorporates teacher intentions into the learning process, leading to provably faster convergence and improved performance across various tasks.

Contribution

It proposes a gradient-based iterative learning method that models teacher awareness, providing theoretical proofs and extensive experiments demonstrating its effectiveness.

Findings

Proves that teacher-aware learning converges faster than naive methods.

Demonstrates improved performance in regression, classification, and inverse reinforcement learning.

Shows benefits of modeling human teacher intentions in machine learning.

Abstract

In human pedagogy, teachers and students can interact adaptively to maximize communication efficiency. The teacher adjusts her teaching method for different students, and the student, after getting familiar with the teacher's instruction mechanism, can infer the teacher's intention to learn faster. Recently, the benefits of integrating this cooperative pedagogy into machine concept learning in discrete spaces have been proved by multiple works. However, how cooperative pedagogy can facilitate machine parameter learning hasn't been thoroughly studied. In this paper, we propose a gradient optimization based teacher-aware learner who can incorporate teacher's cooperative intention into the likelihood function and learn provably faster compared with the naive learning algorithms used in previous machine teaching works. We give theoretical proof that the iterative teacher-aware learning (ITAL) process leads to local and global improvements. We then validate our algorithms with extensive experiments on various tasks including regression, classification, and inverse reinforcement learning using synthetic and real data. We also show the advantage of modeling teacher-awareness when agents are learning from human teachers.