Accurate Predictions in Education with Discrete Variational Inference

2025-09-27

AI Summary

Educational inequality stems from uneven access to high-quality tutoring resources, necessitating scalable and affordable AI-powered tutoring systems. A core challenge lies in accurately predicting student response outcomes under low-data regimes. This paper proposes a novel framework based on discrete variational inference that unifies item response theory (IRT), topic-level skill modeling, and collaborative filtering. Remarkably, it achieves state-of-the-art prediction accuracy using only a single latent ability parameter—challenging conventional multidimensional skill modeling paradigms—while relying on minimal assumptions. Evaluated on the largest publicly available mathematics examination dataset, our model attains over 80% accuracy, significantly outperforming classical IRT and matrix factorization baselines. The results establish a new benchmark for AI-driven educational applications in resource-constrained settings, demonstrating both theoretical elegance and practical efficacy in addressing data-scarce adaptive learning scenarios.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Learning to Search
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

One of the largest drivers of social inequality is unequal access to personal tutoring, with wealthier individuals able to afford it, while the majority cannot. Affordable, effective AI tutors offer a scalable solution. We focus on adaptive learning, predicting whether a student will answer a question correctly, a key component of any effective tutoring system. Yet many platforms struggle to achieve high prediction accuracy, especially in data-sparse settings. To address this, we release the largest open dataset of professionally marked formal mathematics exam responses to date. We introduce a probabilistic modelling framework rooted in Item Response Theory (IRT) that achieves over 80 percent accuracy, setting a new benchmark for mathematics prediction accuracy of formal exam papers. Extending this, our collaborative filtering models incorporate topic-level skill profiles, but reveal a surprising and educationally significant finding, a single latent ability parameter alone is needed to achieve the maximum predictive accuracy. Our main contribution though is deriving and implementing a novel discrete variational inference framework, achieving our highest prediction accuracy in low-data settings and outperforming all classical IRT and matrix factorisation baselines.

Problem

• Predicting student performance on math exams using probabilistic modeling
• Addressing data scarcity in educational prediction through variational inference
• Determining optimal skill representation for accurate educational assessment

Innovation

• Novel discrete variational inference framework implementation
• Single latent ability parameter maximizes predictive accuracy
• Probabilistic modeling rooted in Item Response Theory