TY - GEN
T1 - No RL, No Simulation
T2 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021
AU - Hahn, Meera
AU - Chaplot, Devendra
AU - Tulsiani, Shubham
AU - Mukadam, Mustafa
AU - Rehg, James M.
AU - Gupta, Abhinav
N1 - Funding Information:
The authors would like to thank Saurabh Gupta for the discussions. We would also like to thank the Gibson and RealEstate10K dataset authors for sharing their datasets for scientific research.
Publisher Copyright:
© 2021 Neural information processing systems foundation. All rights reserved.
PY - 2021
Y1 - 2021
N2 - Most prior methods for learning navigation policies require access to simulation environments, as they need online policy interaction and rely on ground-truth maps for rewards. However, building simulators is expensive (requires manual effort for each and every scene) and creates challenges in transferring learned policies to robotic platforms in the real-world, due to the sim-to-real domain gap. In this paper, we pose a simple question: Do we really need active interaction, ground-truth maps or even reinforcement-learning (RL) in order to solve the image-goal navigation task? We propose a self-supervised approach to learn to navigate from only passive videos of roaming. Our approach, No RL, No Simulator (NRNS), is simple and scalable, yet highly effective. NRNS outperforms RL-based formulations by a significant margin. We present NRNS as a strong baseline for any future image-based navigation tasks that use RL or Simulation.
AB - Most prior methods for learning navigation policies require access to simulation environments, as they need online policy interaction and rely on ground-truth maps for rewards. However, building simulators is expensive (requires manual effort for each and every scene) and creates challenges in transferring learned policies to robotic platforms in the real-world, due to the sim-to-real domain gap. In this paper, we pose a simple question: Do we really need active interaction, ground-truth maps or even reinforcement-learning (RL) in order to solve the image-goal navigation task? We propose a self-supervised approach to learn to navigate from only passive videos of roaming. Our approach, No RL, No Simulator (NRNS), is simple and scalable, yet highly effective. NRNS outperforms RL-based formulations by a significant margin. We present NRNS as a strong baseline for any future image-based navigation tasks that use RL or Simulation.
UR - http://www.scopus.com/inward/record.url?scp=85124643067&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85124643067&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85124643067
T3 - Advances in Neural Information Processing Systems
SP - 26661
EP - 26673
BT - Advances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021
A2 - Ranzato, Marc'Aurelio
A2 - Beygelzimer, Alina
A2 - Dauphin, Yann
A2 - Liang, Percy S.
A2 - Wortman Vaughan, Jenn
PB - Neural information processing systems foundation
Y2 - 6 December 2021 through 14 December 2021
ER -