AI Paper Atlas

Research map / Time series / Causal and generative time series

End-to-End Probabilistic Framework for Learning with Hard Constraints

Utkarsh Utkarsh, Danielle C. Maddix, Ruijun Ma, Michael W. Mahoney, Bernie Wang

ICLR 2026 · Time series › Causal and generative time series · approach: Probabilistic and conformal forecasting

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
train end-to-end with the hard constraint rather than apply it only at inference (ablation)ProbHardE2E vs ProbConserv · Nonlinear PME m in [4,5]MSE x1e-6: 276.7 -> 1.59Improved
use CRPS loss instead of NLL (ablation)ProbHardE2E-Or · Advection PDECRPS x1e-3: 8.669 -> 2.94Improved
use CRPS loss instead of NLL (ablation)HardC · PME PDECRPS x1e-3: 1.667 -> 1.785Hurt
use CRPS loss instead of NLL (ablation)ProbHardE2E and baselines · PDE datasetsMSEMixed
orthogonal instead of oblique projection on easy and medium PDEs (ablation)ProbHardE2E · Heat, PME, AdvectionCRPS: 10-30% lower CRPSImproved
oblique instead of orthogonal projection on hard Stefan problem (ablation)ProbHardE2E · Stefan PDECRPS x1e-3: 14.147 -> 7.52Improved
orthogonal instead of oblique projection on hierarchical forecasting (ablation)ProbHardE2E · LabourCRPS x1e-3: 36.1 -> 28.6Improved
orthogonal instead of oblique projection (ablation)ProbHardE2E · Tourism-L, Traffic, WikiCRPS x1e-3: 155.2 -> 156.4 (Tourism-L)Hurt
sampling-free closed-form CRPS with projection on distribution parameters instead of 100 posterior samples (ablation)ProbHardE2E · Hierarchical forecasting and PDE training timeepoch time: 3.3-4.6x increase with samplingImproved
add DPPL constraint projection to unconstrained backbone (ablation)VarianceNO · Nonlinear PMEconstraint error: 0.020 -> 0Improved
add convex TVD inequality constraintProbHardE2E · PME shockshock location and negative samplesImproved
soft penalty constraints instead of hard (ablation)SoftC · PDE datasetsconstraint error: 0.852 (Heat) vs 0Hurt

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