MC-CPO: Mastery-Conditioned Constrained Policy Optimization

TL;DR

This paper introduces MC-CPO, a reinforcement learning algorithm that incorporates pedagogical safety constraints to reduce reward hacking in adaptive tutoring systems, ensuring safer and more effective learning outcomes.

Contribution

The paper presents a novel mastery-conditioned constrained policy optimization algorithm that embeds pedagogical structure into the feasible action space for safer reinforcement learning in education.

Findings

MC-CPO satisfies safety constraints within tolerance across experiments.

It reduces safety costs compared to baseline methods.

It significantly lowers the Reward Hacking Severity Index (RHSI).

Abstract

Engagement-optimized adaptive tutoring systems may prioritize short-term behavioral signals over sustained learning outcomes, creating structural incentives for reward hacking in reinforcement learning policies. We formalize this challenge as a constrained Markov decision process (CMDP) with mastery-conditioned feasibility, in which pedagogical safety constraints dynamically restrict admissible actions according to learner mastery and prerequisite structure. We introduce Mastery-Conditioned Constrained Policy Optimization (MC-CPO), a two-timescale primal-dual algorithm that integrates structural action masking with constrained policy optimization. In the tabular regime, we establish feasibility preservation and convergence to stationary feasible points under standard stochastic approximation conditions and derive a safety gap result showing that optimization within the mastery-conditioned feasible set can strictly dominate post-hoc filtering under identical safety budgets. Empirical validation is conducted in minimal and extended tabular environments and in a neural tutoring setting. Across 10 random seeds and one million training steps in the neural regime, MC-CPO satisfies constraint budgets within tolerance, reduces discounted safety costs relative to unconstrained and reward-shaped baselines, and substantially lowers the Reward Hacking Severity Index (RHSI). These results indicate that embedding pedagogical structure directly into the feasible action space provides a principled foundation for mitigating reward hacking in instructional reinforcement learning systems.