AI Paper Atlas

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

Inference from Quantized Data via Normal Variance-Mean Mixtures

Chenyu Gao, Zhexian Yang, Ziping Zhao

ICML 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Likelihood-free and graphical model estimation

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
replace Gaussian with symmetric GH distribution in multi-bit quantized matrix completionmulti-bit ECM · Quantized matrix completion, MovieLens 1M (40% train)RMSE: 0.9151 -> 0.8908Improved
replace Gaussian with t distribution in multi-bit quantized matrix completionmulti-bit ECM · Quantized matrix completion, MovieLens 1M (40% train)RMSE: 0.9151 -> 0.9091Improved
replace Gaussian with symmetric GH distributionmulti-bit ECM · Quantized matrix completion, MovieLens 1M (40% train)accuracy: 0.4453 -> 0.4503Improved
use multi-bit instead of 1-bit quantization model with GaussianGaussian ECM · Quantized matrix completion, MovieLens 1M (40% train)RMSE: 0.9659 -> 0.9151Improved
use multi-bit GH instead of SVD baselinematrix completion · Quantized matrix completion, MovieLens 1M (40% train)RMSE: 0.9148 -> 0.8908Improved
use multi-bit GH instead of Gaussian: running timemulti-bit ECM · Quantized matrix completion, MovieLens 1M (40% train)time (s): 90.84 -> 272.18Hurt
replace Gaussian with GH distribution in 1-bit compressive sensing1-bit ECM · 1-bit compressive sensing, SNR = -5 dBcosine similarity: 0.5487 -> 0.7705Improved
replace Gaussian with GH distribution in 1-bit compressive sensing1-bit ECM · 1-bit compressive sensing, SNR = -10 dBcosine similarity: 0.2896 -> 0.5386Improved
replace Gaussian with GH distribution in 1-bit compressive sensing1-bit ECM · 1-bit compressive sensing, SNR = 0 dBcosine similarity: 0.8220 -> 0.8906Improved
replace Gaussian with t distribution in 1-bit compressive sensing1-bit ECM · 1-bit compressive sensing, SNR = -10 dBcosine similarity: 0.2896 -> 0.3031No clear effect
use heavy-tailed or skewed NVMM models instead of OLS and Gaussian EMquantized linear regression · synthetic, t and log-normal noiseMSEImproved

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