AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Drawing out of Distribution with Neuro-Symbolic Generative Models

Yichao Liang, Joshua B. Tenenbaum, Tuan Anh Le, Siddharth N

NeurIPS 2022 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Zero-shot robustness and transfer · 4 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 differentiable spline renderer with a neural decoder (remove explicit spline decoder) (ablation)DooD · out-of-distribution reconstruction (trained on MNIST)reconstruction qualityHurt
remove execution-guided inference (no canvas-so-far conditioning) (ablation)DooD · out-of-distribution reconstruction (trained on MNIST)stroke-count selectivity and reconstructionHurt
use DooD instead of AIRstroke-based generative model · Omniglot 20-way one-shot classificationaccuracy: 14.5\% -> 73.5\%Improved
use DooD instead of GNSunconditional generator · Omniglot unconditional generationFID: 0.133 -> 0.051Improved
use DooD stroke latents instead of AIR part representationsk-means clustering of inferred strokes · stroke clusteringSilhouette Coefficient: 0.11 -> 0.21Improved
train on MNIST or Omniglot as source dataset instead of other datasetsDooD and AIR · zero-shot transfer log marginal likelihoodlog marginal likelihoodHurt
train and test on same dataset (diagonal) instead of transferringDooD and AIR · zero-shot transfer log marginal likelihoodlog marginal likelihoodImproved
use Gaussian instead of Laplace likelihood for Omniglot-trained AIRAIR · Omniglot-trained transferreconstructionImproved

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