NeurIPS Competition Instructions and Guide: Causal Insights for Learning Paths in Education

TL;DR

This paper presents a NeurIPS competition focused on causal inference in educational time-series data, aiming to improve personalized learning by identifying causal relationships and predicting learning impacts.

Contribution

It introduces a challenge for causal discovery and impact prediction in education, utilizing synthetic and real-world data to advance machine learning applications in edtech.

Findings

Participants will develop models to identify causal relationships between learning constructs.

Models will predict the impact of learning one construct on others.

The competition evaluates methods using synthetic and real-world A/B test data.

Abstract

In this competition, participants will address two fundamental causal challenges in machine learning in the context of education using time-series data. The first is to identify the causal relationships between different constructs, where a construct is defined as the smallest element of learning. The second challenge is to predict the impact of learning one construct on the ability to answer questions on other constructs. Addressing these challenges will enable optimisation of students' knowledge acquisition, which can be deployed in a real edtech solution impacting millions of students. Participants will run these tasks in an idealised environment with synthetic data and a real-world scenario with evaluation data collected from a series of A/B tests.