ML Newsnew | past | best | rss | submit about
Learning Cognitive Models Using Neural Networks [pdf] (arxiv.org)
65 points by stablemap on Jun 23, 2018 | hide | past | pdf | 1 comment on HN

In plain words: A neural network learns which skills a task demands by studying the tasks themselves, needing no student data and little hand-crafted knowledge. It found accurate skill models in messy domains and estimated skill difficulty and learning speed that closely matched estimates built from student data.

Abstract · Learning Cognitive Models using Neural Networks

A cognitive model of human learning provides information about skills a learner must acquire to perform accurately in a task domain. Cognitive models of learning are not only of scientific interest, but are also valuable in adaptive online tutoring systems. A more accurate model yields more effective tutoring through better instructional decisions. Prior methods of automated cognitive model discovery have typically focused on well-structured domains, relied on student performance data or involved substantial human knowledge engineering. In this paper, we propose Cognitive Representation Learner (CogRL), a novel framework to learn accurate cognitive models in ill-structured domains with no data and little to no human knowledge engineering. Our contribution is two-fold: firstly, we show that representations learnt using CogRL can be used for accurate automatic cognitive model discovery without using any student performance data in several ill-structured domains: Rumble Blocks, Chinese Character, and Article Selection. This is especially effective and useful in domains where an accurate human-authored cognitive model is unavailable or authoring a cognitive model is difficult. Secondly, for domains where a cognitive model is available, we show that representations learned through CogRL can be used to get accurate estimates of skill difficulty and learning rate parameters without using any student performance data. These estimates are shown to highly correlate with estimates using student performance data on an Article Selection dataset.

Devendra Singh Chaplot, Christopher MacLellan, Ruslan Salakhutdinov, Kenneth Koedinger
arXiv:1806.08065 · cs.LG, cs.AI, stat.ML · submitted Jun 21, 2018
abstract · pdf · html

add comment on HN

cs702 on Jun 23, 2018 [–]
"As shown in Figure 1, the neural architecture, which is domain-specific, is connected to a fixed size pre-output layer, which will serve as the representations for corresponding problems. The preoutput layer is in turn connected to the output layer which predicts the correct answer for the given input problem. After training the architecture on the problems in the tutor, we use the trained model to compute the representations vectors in the pre-output layer for each problem. These representations are thresholded at 0.95 and used as columns of the estimated Q-matrix. In other words, each dimension of the learned representation constitutes a Knowledge Component in the predicted cognitive model. This cognitive model is evaluated by fitting an Additive Factors Model using the student performance data."

In other words, a fixed-size preoutput layer added to a range of different domain-specific architectures constructs a task-specific Q-matrix as good as or better than those hand-designed by human experts (from human learning data), simply by training the neural net to learn to perform the task. A "Q-matrix" is a matrix describing relations of questions and concepts required for their understanding in a domain of human knowledge.

We can now use neural nets to create Q-matrices in domains where an accurate human-authored cognitive model is unavailable or authoring a cognitive model is difficult.

Neat.

reply
About | RSS | RSS (all) | HN arXiv