AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Generalized Belief Transport

Junqi Wang, PEI WANG, Patrick Shafto

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Probabilistic embeddings and relational inference · 0 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
learner estimates eta empirically from current data (update) instead of having access to the true hypothesis (exact) (ablation)GBT learner · 10x10 and 5x3 matrices, convergence roundsepisodes reaching 0.999 threshold by roundImproved
increase number of data outcomes (dimension of D) with fixed number of hypothesesGBT learner · m x 10 rectangular matrices, m in 5..100convergence speedImproved
scale square problem size m from 10 to 100GBT learner · m x m matricesconvergenceNo clear effect
vary eps_eta from 0 to infinity with eps_P=1, eps_theta=infinity (ablation)GBT learner · 5x5 problem, 400,000 episodeslimit distribution over hypothesesMixed
set eps_theta=0 so the column-sum constraint no longer affects the transport plan (ablation)GBT learner · theory (Proposition 14)posterior convergenceHurt
replace sudden Bayesian-to-discriminative jump with gradual interpolation (Orange path) (ablation)GBT learner · 4x4 matrix, 10 samplesconclusiveness (mean): 0.9406 -> 0.9964Improved
replace sudden jump with parabola interpolation (Red path) (ablation)GBT learner · 4x4 matrix, 10 samplesconclusiveness (mean): 0.9406 -> 0.9834Improved
replace sudden jump with gradual interpolation (Orange path) (ablation)GBT learner · 4x4 matrix, 10 samplesposterior entropy (mean): 0.1261 -> 0.0079Improved
replace sudden jump with parabola interpolation (Red path) (ablation)GBT learner · 4x4 matrix, 10 samplesposterior entropy (mean): 0.1261 -> 0.0388Improved
use general GBT (eps in cube interior) instead of Bayesian or SCBI learner in periodically changing worldGBT learner · 3x3 matrix, period 18, 400 stepsdetection of periodicity in posteriorImproved
vary radius and shape of the true hypothesis pathGBT learner eps=(1,10,10) · period 20, 300 stepsposterior to hypothesis path area ratio: 0.1620 -> 0.1616No clear effect

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