Salvato in:
Dettagli Bibliografici
Autori principali: Cheng, Weihua, Du, Hanwen, Li, Chunxiao, Ni, Ersheng, Tan, Liangdi, Xu, Tianqi, Ni, Yongxin
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2501.05415
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909462561816576
author Cheng, Weihua
Du, Hanwen
Li, Chunxiao
Ni, Ersheng
Tan, Liangdi
Xu, Tianqi
Ni, Yongxin
author_facet Cheng, Weihua
Du, Hanwen
Li, Chunxiao
Ni, Ersheng
Tan, Liangdi
Xu, Tianqi
Ni, Yongxin
contents Knowledge Tracing (KT) is crucial in education assessment, which focuses on depicting students' learning states and assessing students' mastery of subjects. With the rise of modern online learning platforms, particularly massive open online courses (MOOCs), an abundance of interaction data has greatly advanced the development of the KT technology. Previous research commonly adopts deterministic representation to capture students' knowledge states, which neglects the uncertainty during student interactions and thus fails to model the true knowledge state in learning process. In light of this, we propose an Uncertainty-Aware Knowledge Tracing model (UKT) which employs stochastic distribution embeddings to represent the uncertainty in student interactions, with a Wasserstein self-attention mechanism designed to capture the transition of state distribution in student learning behaviors. Additionally, we introduce the aleatory uncertainty-aware contrastive learning loss, which strengthens the model's robustness towards different types of uncertainties. Extensive experiments on six real-world datasets demonstrate that UKT not only significantly surpasses existing deep learning-based models in KT prediction, but also shows unique advantages in handling the uncertainty of student interactions.
format Preprint
id arxiv_https___arxiv_org_abs_2501_05415
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Uncertainty-aware Knowledge Tracing
Cheng, Weihua
Du, Hanwen
Li, Chunxiao
Ni, Ersheng
Tan, Liangdi
Xu, Tianqi
Ni, Yongxin
Machine Learning
Knowledge Tracing (KT) is crucial in education assessment, which focuses on depicting students' learning states and assessing students' mastery of subjects. With the rise of modern online learning platforms, particularly massive open online courses (MOOCs), an abundance of interaction data has greatly advanced the development of the KT technology. Previous research commonly adopts deterministic representation to capture students' knowledge states, which neglects the uncertainty during student interactions and thus fails to model the true knowledge state in learning process. In light of this, we propose an Uncertainty-Aware Knowledge Tracing model (UKT) which employs stochastic distribution embeddings to represent the uncertainty in student interactions, with a Wasserstein self-attention mechanism designed to capture the transition of state distribution in student learning behaviors. Additionally, we introduce the aleatory uncertainty-aware contrastive learning loss, which strengthens the model's robustness towards different types of uncertainties. Extensive experiments on six real-world datasets demonstrate that UKT not only significantly surpasses existing deep learning-based models in KT prediction, but also shows unique advantages in handling the uncertainty of student interactions.
title Uncertainty-aware Knowledge Tracing
topic Machine Learning
url https://arxiv.org/abs/2501.05415