RL-CycleGAN: Reinforcement Learning Aware Simulation-to-Real
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Deep neural network based reinforcement learning (RL) can learn appropriate visual representations for complex tasks like vision-based robotic grasping without the need for manually engineering or prior learning a perception system. However, data for RL is collected via running an agent in the desired environment, and for applications like robotics, running a robot in the real world may be extremely costly and time consuming. Simulated training offers an appealing alternative, but ensuring that policies trained in simulation can transfer effectively into the real world requires additional machinery. Simulations may not match reality, and typically bridging the simulation-to-reality gap requires domain knowledge and task-specific engineering. We can automate this process by employing generative models to translate simulated images into realistic ones. However, this sort of translation is typically task-agnostic, in that the translated images may not preserve all features that are relevant to the task. In this paper, we introduce the RL-scene consistency loss for image translation, which ensures that the translation operation is invariant with respect to the Q-values associated with the image. This allows us to learn a task-aware translation. Incorporating this loss into unsupervised domain translation, we obtain the RL-CycleGAN, a new approach for simulation-to-real-world transfer for reinforcement learning. In evaluations of RL-CycleGAN on two vision-based robotics grasping tasks, we show that RL-CycleGAN offers a substantial improvement over a number of prior methods for sim-to-real transfer, attaining excellent real-world performance with only a modest number of real-world observations.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Unpaired image-to-image translation using cycle-consistent adversarial networks
Zhu, J.-Y., Park, T., Isola, P., & Efros, A. A. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.244 · Cited by 23,602
Watkins, C. J. C. H., & Dayan, P. (1992) · Machine Learning · doi:10.1007/bf00992698 · Cited by 9,083
Learning Transferable Features with Deep Adaptation Networks
Long, M., Cao, Y., Wang, J., & Jordan, M. I. (2015) · arXiv · doi:10.48550/arxiv.1502.02791 · Cited by 4,505
Generative Adversarial Networks
Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1406.2661 · Cited by 4,477
Large Scale GAN Training for High Fidelity Natural Image Synthesis
Brock, A., Donahue, J., & Simonyan, K. (2018) · arXiv · doi:10.48550/arxiv.1809.11096 · Cited by 3,717
Addressing Function Approximation Error in Actor-Critic Methods
Fujimoto, S., van Hoof, H., & Meger, D. (2018) · arXiv · doi:10.48550/arxiv.1802.09477 · Cited by 3,650 · cited 2×
“To estimate V ( s (cid:48) ) , we use Clipped Double-Q Learning [11].”
MethodSelf-Attention Generative Adversarial Networks
Zhang, H., Goodfellow, I., Metaxas, D., & Odena, A. (2018) · arXiv · doi:10.48550/arxiv.1805.08318 · Cited by 3,353 · cited 1×
“4,16,15]. Pixel-level adaptation is an especially challenging image-translation task when we do not have paired data. Prior techniques tackle this problem using generative adversarial networks (GANs) [13,36,5], conditioning the GAN generator on the simulated image’s pixels. Our technique is based on the CycleGAN pixel…”
MethodDomain randomization for transferring deep neural networks from simulation to the real world
Tobin, J., Fong, R., Ray, A., Schneider, J., Zaremba, W., & Abbeel, P. (2017) · 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8202133 · Cited by 3,124 · cited 2×
“training images automatically [3] to randomizing the simulation environment in the hopes that such randomization will improve the transferability of the learned representations [27, 29, 18, 24, 17].”
BackgroundCyCADA: Cycle-Consistent Adversarial Domain Adaptation
Hoffman, J., Tzeng, E., Park, T., Zhu, J.-Y., Isola, P., Saenko, K., Efros, A. A., & Darrell, T. (2017) · arXiv · doi:10.48550/arxiv.1711.03213 · Cited by 2,656 · cited 1×
“Prior methods can be split into feature-level adaptation, where they learn domain-invariant features [14, 6, 22, 12], and pixel-level adaptation, where they condition on pixels from a source image and re-style it look like an image from the target domain [4, 31, 16, 15].”
MethodUnsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks
Bousmalis, K., Silberman, N., Dohan, D., Erhan, D., & Krishnan, D. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.18 · Cited by 1,603 · cited 1×
“Prior methods can be split into feature-level adaptation, where they learn domain-invariant features [14, 6, 23, 12], and pixel-level adaptation, where they condition on pixels from a source image and re-style it look like an image from the target domain [4, 34, 16, 15].”
MethodData-Driven Grasp Synthesis—A Survey
Bohg, J., Morales, A., Asfour, T., & Kragic, D. (2014) · IEEE Transactions on Robotics · doi:10.1109/tro.2013.2289018 · Cited by 929
CAD2RL: Real Single-Image Flight Without a Single Real Image
Sadeghi, F., & Levine, S. (2017) · Robotics: Science and Systems XIII · doi:10.15607/rss.2017.xiii.034 · Cited by 618
Self-citationLearning Synergies Between Pushing and Grasping with Self-Supervised Deep Reinforcement Learning
Zeng, A., Song, S., Welker, S., Lee, J., Rodriguez, A., & Funkhouser, T. (2018) · 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8593986 · Cited by 606
James, S., Wohlhart, P., Kalakrishnan, M., Kalashnikov, D., Irpan, A., Ibarz, J., Levine, S., Hadsell, R., & Bousmalis, K. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.01291 · Cited by 420 · cited 5×
“One related pixel-level method is RCAN [19], which learns a model mapping images from randomized simulations to a canonical simulation.”
MethodInfluentialSelf-citationTransferring End-to-End Visuomotor Control from Simulation to Real World for a Multi-Stage Task
James, S., Davison, A. J., & Johns, E. (2017) · arXiv · doi:10.48550/arxiv.1707.02267 · Cited by 208 · cited 2×
“training images automatically [3] to randomizing the simulation environment in the hopes that such randomization will improve the transferability of the learned representations [27, 29, 18, 24, 17].”
Background
Cited by
Contrastive Learning for Unpaired Image-to-Image Translation
Park, T., Efros, A. A., Zhang, R., & Zhu, J.-Y. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58545-7_19 · Cited by 1,442
How to train your robot with deep reinforcement learning: Lessons we have learned
Ibarz, J., Tan, J., Finn, C., Kalakrishnan, M., Pastor, P., & Levine, S. (2021) · The International Journal of Robotics Research · doi:10.1177/0278364920987859 · Cited by 570
Self-citationGoogle Scanned Objects: A High-Quality Dataset of 3D Scanned Household Items
Downs, L., Francis, A., Koenig, N., Kinman, B., Hickman, R., Reymann, K., McHugh, T. B., & Vanhoucke, V. (2022) · 2022 International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra46639.2022.9811809 · Cited by 377
Advances and Challenges in Meta-Learning: A Technical Review
Vettoruzzo, A., Bouguelia, M.-R., Vanschoren, J., Rögnvaldsson, T., & Santosh, K. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3357847 · Cited by 314
ReLMoGen: Leveraging Motion Generation in Reinforcement Learning for Mobile Manipulation
Xia, F., Li, C., Martín-Martín, R., Litany, O., Toshev, A., & Savarese, S. (2020) · arXiv · doi:10.48550/arxiv.2008.07792 · Cited by 33
Metadata
- Authors
- Rao, K., Harris, C., Irpan, A., Levine, S., Ibarz, J., & Khansari, M.
- Venue
- 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2020-06
- Pages
- 11154-11163
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W3035756007
- Cited by
- 160
- References
- 15
- FWCI
- 10.28
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.