AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Cross-View Policy Learning for Street Navigation

Ang Li, Huiyi Hu, Piotr Mirowski, Mehrdad Farajtabar

ICCV 2019 · Reinforcement learning › Visual and representation RL · approach: Vision-based robot RL · 40 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 cross-view policy learning with aerial-view adaptation (vs single-view agent)street navigation agent · StreetAir NYU target regionzero-shot reward: 5 -> 29Improved
add cross-view policy learning with aerial-view adaptation (vs single-view agent)street navigation agent · StreetAir NYU target regionzero-shot success rate: 10.5% -> 34.5%Improved
add cross-view policy learning (vs single-view)street navigation agent · StreetAir NYU, reward at 100M stepsreward: 50 -> 190Improved
add cross-view policy learning (vs single-view)street navigation agent · StreetAir NYU, reward at 200M stepsreward: 200 -> 280Improved
add cross-view policy learning (vs single-view)street navigation agent · StreetAir NYU, convergenceconvergence reward: 500 -> 580Improved
add heading prediction auxiliary tasksingle-view and cross-view agents · StreetAir NYU, transferreward: approximately 450 reward at step 350MImproved
add heading prediction auxiliary task (gap between cross-view and single-view shrinks)cross-view vs single-view comparison · StreetAir NYU, transferreward gapMixed
add training curriculum (ablation)single-view agent · StreetAir NYU, transferreward at 100M steps: over 50 reward within 100M steps -> 30 less than the one trained with curriculumImproved
add training curriculum (ablation)cross-view agent · StreetAir NYU, transferreward at 200M steps: around 290 at 200M stepsNo clear effect
add aerial-view adaptation stage (ablation)cross-view agent · StreetAir NYU, transferzero-shot reward and learning speedImproved
add aerial-view adaptation stage (ablation)cross-view agent · StreetAir NYU, after many training stepsrewardNo clear effect
add KL policy distillation loss to L2 embedding loss (ablation)cross-view agent · StreetAir NYU, transfer with curriculum and headingrewardImproved

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