AI Paper Atlas

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

Interpretable agent communication from scratch (with a generic visual processor emerging on the side)

Roberto Dessi, Eugene Kharitonov, Marco Baroni

NeurIPS 2021 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Emergent communication in referential games · 9 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 data augmentation (different views for Sender and Receiver) (ablation)Communication game (separate CNNs) · Gaussian blobsgame accuracy (%): 43.4% -> 0.8% (chance)Improved
add data augmentation (ablation)Communication game (separate CNNs) · ILSVRC-2012 val gamegame accuracy (%): 91.2% -> 81.5%Hurt
add data augmentation (ablation)Communication game · ILSVRC-val protocolnMI: 0.50 -> 0.58Improved
share CNN weights between Sender and Receiver (ablation)Communication game with augmentation · ILSVRC-2012 val gamegame accuracy (%): 81.5% -> 82.2%No clear effect
increase training distractors from 1 to 127 (ablation)Game +aug -shared · ILSVRC-2012 val gamegame accuracy (%): 7.3% -> 81.5%Improved
increase training distractors from 63 to 127 (ablation)Game +aug -shared · OOD set gamegame accuracy (%): 66.0% -> 72.0%Improved
reduce vocabulary size from 2048 to 512 (ablation)Game +aug -shared · ILSVRC-2012 val gamegame accuracy (%): 81.5% -> 68.8%Hurt
increase vocabulary size from 2048 to 4096 (ablation)Game +aug -shared · ILSVRC-2012 val gamegame accuracy (%): 81.5% -> 77.7%No clear effect
use game-trained Sender CNN features instead of SimCLR featureslinear evaluation · ILSVRC-2012 linear evaltop-1 acc: 60.6% -> 60.2%No clear effect
train communication game instead of discretizing SimCLRreferential game · OOD set gamegame accuracy (%): 47.4% -> 72.0%Improved
apply augmentation at test time to a model trained without augmentation (ablation)Game -aug -shared · ILSVRC-2012 val gamegame accuracy (%): 91.2% -> 1.9%Hurt

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