Research map / Probabilistic and scientific modeling / Molecules and proteins
Uprooting and Rerooting Higher-Order Graphical Models
Mark Rowland, Adrian Weller
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 changed | Where | Measured | Effect |
|---|---|---|---|
| reroot at the K-heuristic variable instead of the original root | approximate inference on higher-order binary models · random complete hypergraphs and toroidal grids, log Z | average absolute error | Improved |
| reroot at the G-heuristic variable instead of the original root | approximate inference on higher-order binary models · random complete hypergraphs and toroidal grids, log Z | average absolute error | Improved |
| use K or G heuristic instead of maxtW on pairwise models (ablation) | rerooting heuristic · pairwise models | inference error | Improved |
| increase strength of 1-potentials (larger Wmax) (ablation) | rerooting benefit · random models, log Z | gain in accuracy from rerooting | Hurt |
| increase strength of pure k-potentials for k>1 (ablation) | rerooting benefit · random models, log Z | gain in accuracy from rerooting | Improved |
| reroot instead of inferring on the original model | inference running time · all runs | running time | No clear effect |
Closest papers by meaning
- Uprooting and Rerooting Graphical Models (ICML 2016 · similarity 0.94)
- Fixed Points of Belief Propagation—An Analysis via Polynomial Homotopy Continuation (TPAMI 2017 · similarity 0.86)
- Inference in Graphical Models via Semidefinite Programming Hierarchies (NeurIPS 2017 · similarity 0.86)
- Error-Correcting and Verifiable Parallel Inference in Graphical Models (AAAI 2020 · similarity 0.86)
- Towards an Understanding of Stepwise Inference in Transformers: A Synthetic Graph Navigation Model (ICML 2024 · similarity 0.84)
- The Limits of Inference Scaling Through Resampling (ICLR 2026 · similarity 0.83)
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.
