Deep Knowledge Tracing is an implicit dynamic multidimensional item response theory model

TL;DR

This paper reinterprets deep knowledge tracing as an encoder-decoder model, revealing that simpler decoders can outperform the original DKT in predicting student performance across various datasets.

Contribution

It introduces a new perspective on DKT as an encoder-decoder architecture and demonstrates that simpler decoders can achieve better predictive performance.

Findings

Simpler decoders can outperform original DKT models.

Reinterpreting DKT as an encoder-decoder enables new model improvements.

Better performance achieved with fewer parameters.

Abstract

Knowledge tracing consists in predicting the performance of some students on new questions given their performance on previous questions, and can be a prior step to optimizing assessment and learning. Deep knowledge tracing (DKT) is a competitive model for knowledge tracing relying on recurrent neural networks, even if some simpler models may match its performance. However, little is known about why DKT works so well. In this paper, we frame deep knowledge tracing as a encoderdecoder architecture. This viewpoint not only allows us to propose better models in terms of performance, simplicity or expressivity but also opens up promising avenues for future research directions. In particular, we show on several small and large datasets that a simpler decoder, with possibly fewer parameters than the one used by DKT, can predict student performance better.