Learning Program Embeddings to Propagate Feedback on Student Code
This addresses the challenge of scalable feedback for open-ended assignments in large-scale online education, though it appears incremental as it builds on existing embedding and propagation techniques.
The paper tackles the problem of providing feedback on student code in massive online classes by introducing a neural network method to encode programs as linear maps between embedded precondition and postcondition spaces, enabling propagation of human comments to orders of magnitude more submissions in datasets like Code.org Hour of Code and Stanford's CS1 course.
Providing feedback, both assessing final work and giving hints to stuck students, is difficult for open-ended assignments in massive online classes which can range from thousands to millions of students. We introduce a neural network method to encode programs as a linear mapping from an embedded precondition space to an embedded postcondition space and propose an algorithm for feedback at scale using these linear maps as features. We apply our algorithm to assessments from the Code.org Hour of Code and Stanford University's CS1 course, where we propagate human comments on student assignments to orders of magnitude more submissions.