AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Generalizing Tree Probability Estimation via Bayesian Networks

Cheng Zhang, Frederick A Matsen IV

NeurIPS 2018 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Nonparametric and PAC-Bayes Bayesian learning · 17 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
Use SBN-EM-α instead of CCDTree probability estimation · DS1–DS8KL divergence to ground truth: DS1: 0.6027 -> 0.0130; DS2: 0.0218 -> 0.0128; DS3: 0.2074 -> 0.0882; DS4: 0.1952 -> 0.0637; DS5: 1.3272 -> 0.8218; DS6: Improved
Add weak regularization to SBN-EM (ablation)SBN-EM · DS1–DS8KL divergence to ground truth: DS1: 0.0136 -> 0.0130; DS2: 0.0199 -> 0.0128; DS3: 0.1243 -> 0.0882; DS4: 0.0763 -> 0.0637; DS5: 0.8599 -> 0.8218; DS6: Improved
Use EM rather than the simple-average lower-bound estimator (ablation)SBN-SA · DS1–DS8KL divergence to ground truth: DS1: 0.0687 -> 0.0136; DS2: 0.0218 -> 0.0199; DS3: 0.1152 -> 0.1243; DS4: 0.1021 -> 0.0763; DS5: 0.8952 -> 0.8599; DS6: Mixed
Use SBN-EM rather than CCDTree probability estimation · DS1–DS8KL divergence to ground truth: DS1: 0.6027 -> 0.0136; DS2: 0.0218 -> 0.0199; DS3: 0.2074 -> 0.1243; DS4: 0.1952 -> 0.0763; DS5: 1.3272 -> 0.8599; DS6: Improved
Use SRF rather than SBN-EM-αTree probability estimation · DS2KL divergence to ground truth: 0.0128 -> 0.0122Improved
Use SBN-EM-α rather than SRFTree probability estimation · DS1, DS3–DS8KL divergence to ground truth: DS1: 0.0155 -> 0.0130; DS3: 0.3539 -> 0.0882; DS4: 0.5322 -> 0.0637; DS5: 11.5746 -> 0.8218; DS6: 10.0159 -> 0.2786; DS7Improved

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