CoChat

IRIS: Implicit Reinforcement without Interaction at Scale for Learning Control from Offline Robot Manipulation Data

Open access
Mandlekar, Ajay, Ramos, Fabio, Boots, Byron, Savarese, Silvio, Fei-Fei, Li, Garg, Animesh, Fox, Dieter · 2020 · 2020 IEEE International Conference on Robotics and Automation (ICRA)
Cited by 78

The DOI matches its Crossref record; 13 of the 13 references it cites were found in a registry or an index.

Abstract

Learning from offline task demonstrations is a problem of great interest in robotics. For simple short-horizon manipulation tasks with modest variation in task instances, offline learning from a small set of demonstrations can produce controllers that successfully solve the task. However, leveraging a fixed batch of data can be problematic for larger datasets and longer-horizon tasks with greater variations. The data can exhibit substantial diversity and consist of suboptimal solution approaches. In this paper, we propose Implicit Reinforcement without Interaction at Scale (IRIS), a novel framework for learning from large-scale demonstration datasets. IRIS factorizes the control problem into a goal-conditioned low-level controller that imitates short demonstration sequences and a high-level goal selection mechanism that sets goals for the low-level and selectively combines parts of suboptimal solutions leading to more successful task completions. We evaluate IRIS across three datasets, including the RoboTurk Cans dataset collected by humans via crowdsourcing, and show that performant policies can be learned from purely offline learning. Additional results at https://sites.google.com/stanford.edu/iris/.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/icra40945.2020.9196935

Cites

13 of 13 references verified0 citation errors · 0 not checkable

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.

A few free checks a day · papers already verified open instantly · results are public

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

Research works printed but not linked: 26

  1. 10.15607/RSS.2018.XIV.009 · DOI · Paper page
  2. ross, 2014, Reinforcement and imitation learning via interactive no-regret learning, arXiv preprint arXiv 1406 5979
  3. pong, 2018, Temporal difference models: Model-free deep rl for model-based control, arXiv preprint arXiv 1802 09085
  4. 10.1109/IROS.2016.7758091 · DOI · Paper page
  5. 10.1109/ICRA.2013.6630809 · DOI · Paper page
  6. 10.15607/RSS.2018.XIV.012 · DOI
  7. fujimoto, 2018, Off-policy deep reinforcement learning without exploration, arXiv preprint arXiv 1812 02588
  8. gao, 2018, Reinforcement learning from imperfect demonstrations, arXiv preprint arxiv 1802 05807
  9. 10.1109/ROBOT.2009.5152709 · DOI · Paper page
  10. higgins, 2017, Beta-vae: Learning basic visual concepts with a constrained variational framework, ICLRE, 2, 6
  11. 10.1177/0278364912445831 · DOI · Paper page
  12. kingma, 2013, Auto-encoding variational bayes, arXiv preprint arXiv 1312 6114
  13. agarwal, 2019, Striving for simplicity in off-policy deep reinforcement learning, arXiv preprint arXiv 1907 09977
  14. 10.1109/ICRA.2016.7487517 · DOI · Paper page
  15. bhatt, 2019, Crossnorm: Normalization for off-policy td reinforcement learning, arXiv preprint arXiv 1902 05023
  16. dundar, 2020, Unsupervised disentanglement of pose, appearance and background from images and videos, arXiv preprint arXiv 2001 09518
  17. ding, 2019, Goal-conditioned imitation learning, arXiv preprint arXiv 1906 03008
  18. achiam, 2019, Towards characterizing divergence in deep q-learning, arXiv preprint arXiv 1903 00066
  19. fan, 2018, Surreal: Open-source reinforcement learning framework and robot manipulation benchmark, Conference on Robot Learning
  20. abbeel, 2011, Inverse reinforcement learning, Encyclopedia of Machine Learning, 554
  21. 10.1177/0278364918784350 · DOI · Paper page
  22. levine, 2016, Learning hand-eye coordination for robotic grasping with large-scale data collection, ISER, 173
  23. kumar, 2019, Stabilizing off-policy q-learning via bootstrapping error reduction, arXiv preprint arXiv 1906 03008
  24. lynch, 2019, Learning latent plans from play, arXiv preprint arXiv 1903 00066
  25. liu, 2019, Off-policy policy gradient with state distribution correction, arXiv preprint arXiv 1904 01870
  26. 10.15607/RSS.2017.XIII.058 · DOI · Paper page

Cited by

See all 78

Metadata

Authors
Mandlekar, A., Ramos, F., Boots, B., Savarese, S., Fei-Fei, L., Garg, A., & Fox, D.
Venue
2020 IEEE International Conference on Robotics and Automation (ICRA)
Publisher
IEEE
Published
2020-05
Pages
4414-4420
ISSN
1050-4729 (ISSN-L)
OpenAlex
W2986616324
Cited by
78
References
13
FWCI
3.73
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.