Pith. sign in

REVIEW 8 cited by

RRL: Resnet as representation for Reinforcement Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2107.03380 v3 pith:GMUUKXVA submitted 2021-07-07 cs.RO

classification cs.RO
keywords learningbehaviorsdirectlyreinforcementrepresentationresnetstatecomplex
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The ability to autonomously learn behaviors via direct interactions in uninstrumented environments can lead to generalist robots capable of enhancing productivity or providing care in unstructured settings like homes. Such uninstrumented settings warrant operations only using the robot's proprioceptive sensor such as onboard cameras, joint encoders, etc which can be challenging for policy learning owing to the high dimensionality and partial observability issues. We propose RRL: Resnet as representation for Reinforcement Learning -- a straightforward yet effective approach that can learn complex behaviors directly from proprioceptive inputs. RRL fuses features extracted from pre-trained Resnet into the standard reinforcement learning pipeline and delivers results comparable to learning directly from the state. In a simulated dexterous manipulation benchmark, where the state of the art methods fail to make significant progress, RRL delivers contact rich behaviors. The appeal of RRL lies in its simplicity in bringing together progress from the fields of Representation Learning, Imitation Learning, and Reinforcement Learning. Its effectiveness in learning behaviors directly from visual inputs with performance and sample efficiency matching learning directly from the state, even in complex high dimensional domains, is far from obvious.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. MimicDroid: In-Context Learning for Humanoid Robot Manipulation from Human Play Videos

    cs.RO 2025-09 conditional novelty 7.0 of 10

    Trained only on unlabeled human play videos, MimicDroid lets a GR1 humanoid perform new manipulation tasks from one to three demonstration videos, with roughly twice the real-world success of prior video-conditioned methods.

  2. V-Simba: Unleashing the Architectural Potential of RL in Visual Continuous Control

    cs.LG 2026-08 conditional novelty 6.0 of 10

    V-Simba, a visual RL architecture combining layer normalization, weight decay, and a distributional critic, matches or outperforms complex baselines on 29 continuous control tasks while using less compute.

  3. UAD: Unsupervised Affordance Distillation for Generalization in Robotic Manipulation

    cs.RO 2025-06 conditional novelty 6.0 of 10

    UAD distills affordance knowledge from vision-language models and DINOv2 features into a lightweight task-conditioned model that predicts pixel-level manipulation regions and improves few-shot imitation learning gener...

  4. MuST: Multi-Head Skill Transformer for Long-Horizon Dexterous Manipulation with Skill Progress

    cs.RO 2025-02 conditional novelty 6.0 of 10

    MuST adds per-skill action heads and a progress-guided skill selector to the Octo robot policy, improving long-horizon pick-and-pack success from about 32% to 90% in one simulated setting.

  5. SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulation

    cs.RO 2025-01 conditional novelty 6.0 of 10

    SAM2Act reports 86.8% average success across 18 RLBench tasks, and the memory variant SAM2Act+ reaches 94.3% on the new MemoryBench tasks.

  6. When Should We Prefer State-to-Visual DAgger Over Visual Reinforcement Learning?

    cs.CV 2024-12 conditional novelty 6.0 of 10

    State-to-Visual DAgger outperforms visual RL on hard manipulation tasks and is more stable and faster in wall-clock time, but offers little sample-efficiency benefit on easy tasks.

  7. 4D Visual Pre-training for Robot Learning

    cs.CV 2025-08 conditional novelty 5.0 of 10

    A next-frame point-cloud diffusion pre-training method (FVP) improves DP3 and RDT-1B manipulation success rates on the paper's own tasks.

  8. Efficient Reinforcement Learning Through Adaptively Pretrained Visual Encoder

    cs.CV 2025-02 conditional novelty 5.0 of 10

    APE pretrains a ResNet18 encoder with adaptively selected augmentations and freezes its early layers during policy learning, improving sample efficiency of DreamerV3 and DrQ-v2 on several visual RL benchmarks.

Pith tools