AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Amortized Monte Carlo Integration

Adam Golinski, Frank Wood, Tom Rainforth

ICML 2019 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Importance sampling and quadrature · 2 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 three separately trained proposals to estimate positive, negative, and normalization termsSNIS-style amortized expectation estimation · 1D Gaussian tail integralReMSE: about 10^3 more samples to achieve the same relative errorImproved
Use three separately trained proposals to estimate positive, negative, and normalization termsSNIS-style amortized expectation estimation · 5D Gaussian tail integralReMSEImproved
Use three separately trained proposals to estimate positive, negative, and normalization termsSNIS-style amortized expectation estimation · Cancer treatment planningReMSE: N=10^4; N=2Improved
Use three separately trained proposals instead of the mixture proposal in SNISSNIS q_m · 1D Gaussian tail integralReMSE: about 10^3 more samples to achieve the same relative errorImproved
Use three separately trained proposals instead of the mixture proposal in SNISSNIS q_m · 5D Gaussian tail integralReMSEImproved
Use three separately trained proposals instead of the mixture proposal in SNISSNIS q_m · Cancer treatment planningReMSEImproved
Use AMCI rather than the theoretical SNIS error boundSNIS estimator · 5D Gaussian tail integralReMSEHurt
Reuse samples from q_1 to estimate the denominator, setting beta above zero (ablation)AMCI · Tail integral, low-mismatch caseReMSENo clear effect
Reuse samples from q_1 to estimate the numerator, setting alpha below one (ablation)AMCI · Tail integral, low-mismatch caseReMSE: N=M=64Improved
Reuse samples from q_1 to estimate the numerator, setting alpha below one (ablation)AMCI · Tail integral, high-mismatch caseReMSE: N=M=64No clear effect
Train using a rotating synthetic training and validation dataset rather than drawing every batch directly from the model (ablation)AMCI proposals · Training proceduretraining convergenceImproved
Use SNIS q_1 rather than SNIS q_2SNIS · Tail integralReMSEHurt

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