Reproduce “Concept frustration: Aligning human concepts and machine representations” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Enrico Parisini, Christopher J. Soelistyo, Ahab Isaac, Alessandro Barp, Christopher R.S. Banerji
The authors introduce concept frustration, a diagnostic for detecting when an omitted task-relevant concept may distort relationships among concepts in a concept bottleneck model. It compares those relationships using task-aligned, Fisher-information-based geometry. In synthetic experiments, frustration lowered mean CBM accuracy from 0.718 to 0.699; language and vision comparisons showed lower frustration after adding candidate concepts, without significant accuracy gains.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.