AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

Learning to Draw: Emergent Communication through Sketching

Daniela Mihai, Jonathon Hare

NeurIPS 2021 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Emergent communication in referential games · 5 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
Add a perceptual loss matching sketch and photo featuresDrawing communication agents · STL-10 original game; human receiver with 9 distractorshuman communication rate: 8.3\%\ (\pm5.4) -> 38.3\%\ (\pm2.5)Improved
Add perceptual lossDrawing communication agents · STL-10 original gamecommunication success rate: 71.8\%\ (\pm 6.1) -> 69.57\%\ (\pm 2.6)No clear effect
Use an object-oriented game with disjoint classesDrawing communication agents · STL-10human interpretabilityImproved
Increase sender encoding and receiver MLP width from 64 to 1024 (ablation)Drawing communication agents · Caltech-101 OO-game differentcommunication rate: 50.46\%\ (\pm 1.5) -> 64.99\%\ (\pm 1.5)Improved
Replace ImageNet-pretrained weights with Stylized-ImageNet shape-biased weights (ablation)Drawing communication agents · Caltech-101 original gamecommunication success rate: 78.46\%\ (\pm 2.0) -> 77.09\%\ (\pm 1.9)No clear effect
Reduce the sketch from 20 strokes to 5 (ablation)Drawing communication agents · STL-10 original gamecommunication success rate: 69.57\% \ (\pm 2.6) -> 73.41\%\ (\pm 1.6)Improved
Increase stroke count from 20 to 50 (ablation)Drawing communication agents · STL-10 original gamecommunication success rate: 69.57\% \ (\pm 2.6) -> 69.21\% \ (\pm 1.4)Hurt
Use only VGG16 layers 1–4 rather than all five feature layers in perceptual loss (ablation)Drawing communication agents · STL-10 original gamecommunication success rate: 69.57\%\ (\pm 2.6) -> 19.09\%\ (\pm 10.1)Hurt
Use perceptual features only up to VGG16 layer 3 (ablation)Drawing communication agents · STL-10 original gamecommunication success rate: 69.57\%\ (\pm 2.6) -> 1.0\% \ (\pm 0)Hurt
Use only the first or second VGG16 feature block for perceptual loss (ablation)Drawing communication agents · STL-10 original gamehuman interpretabilityHurt
Increase the number of perceptual feature maps beyond the original five (ablation)Drawing communication agents · STL-10 original gamecommunication success rate: 68.4\%\ (\pm 2.0); 69.5\%\ (\pm 1.8); 69.0\% \ (\pm 1.2)No clear effect
Replace photo-matching perceptual loss with an arbitrary dog-image objective (ablation)Drawing communication agents · CelebA original gamehuman interpretabilityHurt

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