AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Mastering Spatial Graph Prediction of Road Networks

Anagnostidis Sotiris, Aurelien Lucchi, Thomas Hofmann

ICCV 2023 · Reinforcement learning › Visual and representation RL · approach: Graph search and compositional goals · 6 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
train with reward and value estimates and tree search instead of cross-entropy autoregressive ARMARM · SpaceNet (ground truth key points)APLS: 0.743 -> 0.778Improved
generate edges in arbitrary order instead of fixed autoregressive orderARM · SpaceNet (ground truth key points)bits per edge, random order: 30.84 -> 4.565Improved
use ours instead of ARM on autoregressive-order perplexityARM · SpaceNet (ground truth key points)bits per edge, autoregressive order: 0.733 -> 4.636Hurt
attend to ViT per-patch image features in the two transformer modules (ablation)Our method · SpaceNettopological metricsNo clear effect
apply model to DeepGlobe without fine-tuning to refine baseline predictionsOur method · DeepGlobeAPLS: 0.6525 -> 0.7398Improved
use smaller simulation depth during training (ablation)Our method · SpaceNettopological metricsNo clear effect
use our method instead of Orientation on SpaceNetOrientation · SpaceNetCCQ quality: 0.6862 -> 0.6632Mixed

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