Fully Adaptive Framework: Neural Computerized Adaptive Testing for Online Education

Abstract

Computerized Adaptive Testing (CAT) refers to an efficient and personalized test mode in online education, aiming to accurately measure student proficiency level on the required subject/domain. The key component of CAT is the "adaptive" question selection algorithm, which automatically selects the best suited question for student based on his/her current estimated proficiency, reducing test length. Existing algorithms rely on some manually designed and pre-fixed informativeness/uncertainty metrics of question for selections, which is labor-intensive and not sufficient for capturing complex relations between students and questions. In this paper, we propose a fully adaptive framework named Neural Computerized Adaptive Testing (NCAT), which formally redefines CAT as a reinforcement learning problem and directly learns selection algorithm from real-world data. Specifically, a bilevel optimization is defined and simplified under CAT's application scenarios to make the algorithm learnable. Furthermore, to address the CAT task effectively, we tackle it as an equivalent reinforcement learning problem and propose an attentive neural policy to model complex non-linear interactions. Extensive experiments on real-world datasets demonstrate the effectiveness and robustness of NCAT compared with several state-of-the-art methods.

A novel computerized adaptive testing framework with decoupled learning selector

“…This category attempts to learn and continuously optimize a question selector from large-scale behavior data, instead of using static question selection algorithms to reduce the error of capacity estimation as much as possible. The representatives of this category include bilevel optimizationbased computerized adaptive testing (BOBCAT) [10] and neural computerized adaptive testing (NCAT) [40]. Both the models redefines the CAT problem as a bilevel optimization problem.…”

“…Specifically, most question selection algorithms are based on predefined criteria, which have certain preferences and cannot effectively capture the complex data characteristics [4,15]. To address the limitations of these heuristic-based question selectors, some experts also attempted to explore question selectors based on learning selection strategies [10] by redefining the CAT problem as a bilevel optimization [10,40] and reinforcement learning problem [5,14]. These learning-based question selectors have shown advantages over the criteria-based question selectors [14].…”

GMOCAT: A Graph-Enhanced Multi-Objective Method for Computerized Adaptive Testing

“…As a question selector, the selection algorithm plays a crucial role in the above CAT process, thus we focus on designing an effective data-driven selection algorithm in this paper. In recent years, data-driven selection algorithms have been proposed from the perspectives of meta learning [14] or Reinforcement Learning (RL) [44]. However, these studies only focus on the quality objective of predicting the student ability, which is insufficient in real-world scenarios [23].…”

“…Firstly, we formalize the CAT procedure as a Multi-Objective Markov decision process (MOMDP) and then introduce a Scalarized Multi-Objective Reinforcement Learning (Scalarized MORL) framework into the CAT setting. Compared with the greedy methods for CAT, the RL framework has been proven to explore more appropriate questions for students from a long-term view [44].…”

“…For example, BOBCAT [14] is a meta learning-based method that couples CDM and selection algorithm together in a bilevel optimization problem. NCAT [44] is a reinforcement learning-based method that utilizes an attention-based DQN to select questions. NCAT also controls the question exposure by sampling from the Boltzmann distribution [15].…”