AI Paper Atlas

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

Learning to Navigate in Cities Without a Map

Piotr Mirowski, Matt Grimes, Mateusz Malinowski, Karl Moritz Hermann, Keith Anderson, Denis Teplyashin, Karen Simonyan, koray kavukcuoglu, Andrew Zisserman, Raia Hadsell

NeurIPS 2018 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Vision-language navigation agents · 52 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
use a dual-LSTM architecture with a locale-specific goal pathwayGoalNav on the NYU courier task · NYU courier taskaverage reward per episodeImproved
add heading prediction to GoalNav (ablation)GoalNav · NYU courier taskaverage reward per episodeImproved
add a skip connection from the vision layers to the policy LSTM (ablation)CityNav · NYU courier taskaverage reward per episodeImproved
remove the goal input (ablation)CityNav · NYU courier taskaverage reward per episodeHurt
add heading prediction to two-LSTM CityNav without a skip connection (ablation)CityNav · NYU courier taskaverage reward per episodeNo clear effect
use CityNav instead of GoalNavNYU courier task · NYU courier taskaverage reward per episodeNo clear effect
add random coin rewards (ablation)CityNav reward training · NYU courier taskgoal rewardNo clear effect
use larger-radius early rewards (ablation)CityNav reward training · NYU courier taskgoal rewardImproved
reduce landmark density (ablation)CityNav goal representation · NYU courier taskaverage reward per episodeHurt
replace landmark goals with normalized latitude and longitude scalar coordinates (ablation)CityNav · NYU courier taskaverage reward per episodeImproved
remove the skip connection in multi-city pre-training and transfer (ablation)MultiCityNav · Wall Street transfergoal rewardImproved
pre-train on more regions before transferring to the target cityMultiCityNav · Wall Street transfergoal rewardImproved

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