AI Paper Atlas

Research map / Learning theory and algorithms / Optimal transport

Efficient Sliced Wasserstein Distance Computation via Adaptive Bayesian Optimization

Manish Acharya, David Hyde

ICLR 2026 · Learning theory and algorithms › Optimal transport · approach: Sliced Wasserstein variants

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 one-shot BO-learned directions (BOSW) instead of QSW/MC for approximating SWBOSW · ShapeNet point-cloud SW approximation errorabsolute errorHurt
periodically refresh BO directions (RBOSW) during gradient flow (ablation)BOSW · Point-cloud interpolation (L=100)W2 at step 100: 6.096 -> 4.155 (at step 100)Improved
periodically refresh BO directions (RBOSW) increases runtime (ablation)BOSW · Point-cloud interpolation (L=100)runtime (s): 3.99 -> 44.45Hurt
seed BO from QSW and restart periodically (ARBOSW) instead of one-shot ABOSW (ablation)ABOSW · Point-cloud interpolation (L=100)W2 at step 500: 0.076 -> 0.003Improved
add QSW seeding with a few BO refinements (ABOSW) to BOSW (ablation)BOSW · Deep point-cloud autoencoderW2 at epoch 100: 10.34 -> 10.27Improved
use restart variants (RBOSW, ARBOSW) instead of non-restart BOSW/ABOSW (ablation)BO selectors · Deep point-cloud autoencoderSW2 at epoch 400: ABOSW 1.81 vs ARBOSW 1.85; BOSW 1.80 vs RBOSW 1.90Hurt
use BO selectors instead of recommended CQSWABOSW · Deep point-cloud autoencoderSW2 at epoch 400: 1.84 -> 1.81Improved
use BO versus QSW on synthetic fitness landscapesBO projection selector · Quadratic landscapebest fitness: 2.69 vs <= 1.0 at L=10Improved
use UCB acquisition instead of EI, Thompson sampling or log-EI (ablation)BO selectors · Point-cloud interpolationconvergenceImproved
set UCB beta to 0.7 versus other values (ablation)BO selectors · Point-cloud interpolationconvergenceImproved
set cosine-similarity cutoff to 0.98 versus 0.9 or 0.95 (ablation)BO selectors · Point-cloud interpolation (L=100)final errorImproved
vary candidate pool size n_c from 512 to 16384 (ablation)BO selectors · Point-cloud interpolationerror after 500 stepsNo 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