Pith. sign in

REVIEW 10 cited by

Learning to Generalize Across Long-Horizon Tasks from Human Demonstrations

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 2003.06085 v2 pith:UMAXXOKZ submitted 2020-03-13 cs.RO cs.AIcs.LG

Learning to Generalize Across Long-Horizon Tasks from Human Demonstrations

classification cs.RO cs.AIcs.LG
keywords learninggeneralizeimitationtrainbehaviorsdemonstrationspoliciesreal
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Imitation learning is an effective and safe technique to train robot policies in the real world because it does not depend on an expensive random exploration process. However, due to the lack of exploration, learning policies that generalize beyond the demonstrated behaviors is still an open challenge. We present a novel imitation learning framework to enable robots to 1) learn complex real world manipulation tasks efficiently from a small number of human demonstrations, and 2) synthesize new behaviors not contained in the collected demonstrations. Our key insight is that multi-task domains often present a latent structure, where demonstrated trajectories for different tasks intersect at common regions of the state space. We present Generalization Through Imitation (GTI), a two-stage offline imitation learning algorithm that exploits this intersecting structure to train goal-directed policies that generalize to unseen start and goal state combinations. In the first stage of GTI, we train a stochastic policy that leverages trajectory intersections to have the capacity to compose behaviors from different demonstration trajectories together. In the second stage of GTI, we collect a small set of rollouts from the unconditioned stochastic policy of the first stage, and train a goal-directed agent to generalize to novel start and goal configurations. We validate GTI in both simulated domains and a challenging long-horizon robotic manipulation domain in the real world. Additional results and videos are available at https://sites.google.com/view/gti2020/ .

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 10 Pith papers

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

  1. RoboInter1.5: A Holistic Intermediate Representation Suite for Embodied World Modeling and Robotic Manipulation

    cs.RO 2026-07 conditional novelty 6.0

    Dense per-frame intermediate representations (traces, masks, grasp poses, subtasks) improve embodied VQA, VLA action generation, and world-model video prediction in the new 230k-episode RoboInter-Data suite.

  2. GHOST: Hierarchical Sub-Goal Policies for Generalizing Robot Manipulation

    cs.RO 2026-06 unverdicted novelty 6.0

    GHOST improves generalization in robot manipulation via hierarchical factorization into 3D sub-goal prediction from RGB-D views and a goal-conditioned low-level controller, enabling human video integration without act...

  3. SID: Sliding into Distribution for Robust Few-Demonstration Manipulation

    cs.RO 2026-05 unverdicted novelty 6.0

    SID achieves approximately 90% success on six real-world manipulation tasks with only two demonstrations under out-of-distribution initializations, with less than 10% performance drop under distractors and disturbances.

  4. When Life Gives You BC, Make Q-functions: Extracting Q-values from Behavior Cloning for On-Robot Reinforcement Learning

    cs.RO 2026-05 unverdicted novelty 6.0

    Q2RL extracts Q-values from a BC policy and applies Q-gating to enable efficient offline-to-online RL, outperforming baselines on D4RL/robomimic tasks and achieving up to 100% success on real-robot manipulation in 1-2 hours.

  5. When Life Gives You BC, Make Q-functions: Extracting Q-values from Behavior Cloning for On-Robot Reinforcement Learning

    cs.RO 2026-05 unverdicted novelty 6.0

    Q2RL extracts Q-functions from BC policies via minimal interactions and applies Q-gating to enable stable offline-to-online RL, outperforming baselines on manipulation benchmarks and achieving up to 100% success on-robot.

  6. RoboMD: Uncovering Robot Vulnerabilities through Semantic Potential Fields

    cs.RO 2024-12 unverdicted novelty 6.0

    A deep RL vulnerability-prediction policy trained in semantic embedding space finds up to 23% more unique robot manipulation failures than vision-language baselines and enables more efficient fine-tuning.

  7. What Matters in Learning from Offline Human Demonstrations for Robot Manipulation

    cs.RO 2021-08 accept novelty 6.0

    A comprehensive benchmark study of offline imitation learning methods on multi-stage robot manipulation tasks identifies key sensitivities to algorithm design, data quality, and stopping criteria while releasing all d...

  8. Imagine to Ensure Safety in Hierarchical Reinforcement Learning

    cs.AI 2026-06 unverdicted novelty 5.0

    Hierarchical safe RL using world-model subgoal generation and imagined rollouts outperforms baselines on long-horizon navigation and manipulation tasks while meeting safety budgets.

  9. MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrations

    cs.RO 2023-10 unverdicted novelty 5.0

    MimicGen creates over 50K robot demonstrations from roughly 200 human ones, allowing imitation learning to achieve strong performance on complex long-horizon tasks like assembly and coffee preparation.

  10. Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems

    cs.LG 2020-05 unverdicted novelty 2.0

    Offline RL promises to extract high-utility policies from static datasets but faces fundamental challenges that current methods only partially address.