AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Learning Graph Structure With A Finite-State Automaton Layer

Daniel Johnson, Hugo Larochelle, Daniel Tarlow

NeurIPS 2020 · Graph learning › Knowledge and temporal graphs · approach: Heterogeneous graph learning · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add GFSA edges halfway through a RATVariable misuse model · Variable misuse, Allfull accuracy: 88.22 -> 89.63Improved
Add GFSA edges before a RATVariable misuse model · Variable misuse, Allfull accuracy: 88.22 -> 89.47Improved
Add GFSA edges halfway through a RAT rather than before itVariable misuse model · Variable misuse, Allfull accuracy: 89.47 -> 89.63Improved
Add GFSA edges halfway through a RATVariable misuse model · Variable misuse, With bug classificationaccuracy: 92.48 -> 94.25Improved
Add GFSA edges halfway through a RATVariable misuse model · Variable misuse, With bug Loc & Repairaccuracy: 88.39 -> 89.93Improved
Add GFSA edges before a RATVariable misuse model · Variable misuse, No bug classificationaccuracy: 93.03 -> 93.10No clear effect
Add random-walk edges before a RATVariable misuse model · Variable misuse, No bug classificationaccuracy: 92.05 -> 93.22Improved
Replace exact trajectory marginals and cross-entropy with single sampled edges trained using REINFORCE (ablation)GFSA program-analysis edge classifier · Last Read, 100 training examples, 1xF1: 98.81 -> 98.48Hurt
Replace exact trajectory marginals and cross-entropy with single sampled edges trained using REINFORCE (ablation)GFSA program-analysis edge classifier · Last Write, 100 training examples, 1xF1: 98.71 -> 98.32Hurt
Replace learned multi-step GFSA policies with a uniform random walk and learned halting probability (ablation)Variable misuse model · Variable misuse, Allfull accuracy: 88.22 -> 88.91No clear effect
Train learned GFSA navigation optionsGrid-world pathfinding · Validation layoutssteps to goal: 40.1 -> 11.5 stepsImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy