Optimizing Coverage and Difficulty in Reinforcement Learning for Quiz Composition

2026-03-29

AI Summary

This study addresses the challenge of balancing knowledge coverage and difficulty control in automated test generation by formulating quiz construction as a sequential decision-making problem. It introduces reinforcement learning—specifically leveraging DQN, SARSA, and A2C/A3C algorithms—to adaptively select questions that jointly satisfy pedagogical objectives regarding coverage and difficulty. The proposed intelligent item selection strategy is evaluated on both synthetic and real-world question banks, demonstrating its ability to efficiently generate high-quality exams aligned with instructor-specified targets. Experimental results further reveal strong generalization capabilities under distribution shifts and in multi-objective settings. User studies corroborate the practical utility of the approach, confirming its potential for real-world educational deployment.

Abstract

Quiz design is a tedious process that teachers undertake to evaluate the acquisition of knowledge by students. Our goal in this paper is to automate quiz composition from a set of multiple choice questions (MCQs). We formalize a generic sequential decision-making problem with the goal of training an agent to compose a quiz that meets the desired topic coverage and difficulty levels. We investigate DQN, SARSA and A2C/A3C, three reinforcement learning solutions to solve our problem. We run extensive experiments on synthetic and real datasets that study the ability of RL to land on the best quiz. Our results reveal subtle differences in agent behavior and in transfer learning with different data distributions and teacher goals. This was supported by our user study, paving the way for automating various teachers' pedagogical goals.

Problem

• quiz composition
• reinforcement learning
• topic coverage
• difficulty optimization
• automated assessment

Innovation

• reinforcement learning
• quiz composition
• topic coverage
• difficulty control
• sequential decision-making