AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Set Prediction without Imposing Structure as Conditional Density Estimation

David W Zhang, Gertjan J. Burghouts, Cees G. M. Snoek

ICLR 2021 · Probabilistic and scientific modeling › Molecules and proteins · approach: Spatiotemporal forecasting and imputation · 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
Use conditional density estimation instead of assignment-based set-loss trainingSet prediction · PolygonsHungarian loss [10^-5]: 1195.0 ± 6.8 / 14.8 ± 5.3 -> 8.8 ± 1.5Improved
Use conditional density estimation instead of assignment-based set-loss trainingSet prediction · DigitsChamfer loss [10^-5]: 2.4 ± 0.3 / 48.0 ± 2.3 -> 1.8 ± 0.1Improved
Use DESP rather than DSPN trained with Chamfer lossPoint-cloud auto-encoding · set MNISTSet size RMSE: 0.300 ± 0.130 / 0.800 ± 0.080 -> 0.002 ± 0.003Improved
Use DESP rather than DSPN trained with assignment lossPoint-cloud auto-encoding · set MNISTChamfer loss [10^-4]: 0.9 ± 0.1 / 2.8 ± 0.8 -> 0.6 ± 0.1Improved
Use DESP rather than DSPN trained with assignment lossPoint-cloud auto-encoding · set MNISTHungarian loss [10^-4]: 1600.2 ± 162.7 / 3.2 ± 0.8 -> 0.6 ± 0.1Improved
Use DESP rather than DSPN with Hungarian trainingObject set prediction · CLEVRAP50: 94.0 ± 0.4 -> 96.2 ± 0.5Improved
Use DESP rather than DSPN with Hungarian trainingObject set prediction · CLEVRAP70: 82.2 ± 1.4 -> 80.9 ± 1.8Hurt
Use DESP rather than DSPN with Hungarian trainingObject set prediction · CLEVRAP80: 58.9 ± 3.4 -> 54.3 ± 3.4Hurt
Use DESP rather than the element-wise DeepSets baselineSubset anomaly detection · CelebA ambiguousF1: 0.47 ± 0.01 -> 0.66 ± 0.00Improved
Use DESP rather than the element-wise DeepSets baselineSubset anomaly detection · CelebA unambiguous + ambiguousF1: 0.63 ± 0.01 -> 0.76 ± 0.01Improved
Add stochastic steps to prediction instead of using fully deterministic gradient descent (ablation)DESP · set MNISTReconstruction performanceImproved
Set the stochastic-step proportion to 0.8 rather than fully deterministic prediction (ablation)DESP · CelebA subset anomaly detectionF1Improved

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