AI Paper Atlas

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

On Fenchel Mini-Max Learning

Chenyang Tao, Liqun Chen, Shuyang Dai, Junya Chen, Ke Bai, Dong Wang, Jianfeng Feng, Wenlian Lu, Georgiy Bobashev, Lawrence Carin

NeurIPS 2019 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Exact inference for probabilistic programs · 10 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
replace VAE ELBO training with FMLVAE · MNISTFID: 24.3 -> 22.7Improved
replace VAE ELBO training with FMLVAE · MS COCOperplexity: 9.5 -> 8.6Improved
replace VAE ELBO training with FMLVAE · EMNLP WMT newsperplexity: 12.5 -> 11.6Improved
use FML instead of DFM (DAE score estimator) for GAN trainingGAN · Cifar10FID: 30.7 -> 30.0Improved
use FML instead of noise contrastive estimationunnormalized models · UCI wine-redtest log-likelihood: 7.29 -> 8.45Improved
use FML instead of NCEunnormalized models · UCI htru2test log-likelihood: 22.05 -> 22.15No clear effect
use FML instead of naive Monte Carlounnormalized models · toy bananaparameter estimation error: 3.46 -> 3.05Improved
use FML instead of NCEunnormalized models · toy waveparameter estimation error: 1.20 -> 1.27Mixed
replace SVGD sampler with FML-trained sampler in soft Q-learningsoft Q-learning · Hopper, Swimmer, ReacherrewardImproved

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