Demo-JEPA enables one-shot cross-embodiment imitation by mapping visual demonstrations to shared latent future trajectories that serve as subgoals for the target agent's own forward dynamics planning.
hub
arXiv preprint arXiv:1805.01954 , year=
19 Pith papers cite this work. Polarity classification is still indexing.
19
Pith papers citing it
abstract
Humans often learn how to perform tasks via imitation: they observe others perform a task, and then very quickly infer the appropriate actions to take based on their observations. While extending this paradigm to autonomous agents is a well-studied problem in general, there are two particular aspects that have largely been overlooked: (1) that the learning is done from observation only (i.e., without explicit action information), and (2) that the learning is typically done very quickly. In this work, we propose a two-phase, autonomous imitation learning technique called behavioral cloning from observation (BCO), that aims to provide improved performance with respect to both of these aspects. First, we allow the agent to acquire experience in a self-supervised fashion. This experience is used to develop a model which is then utilized to learn a particular task by observing an expert perform that task without the knowledge of the specific actions taken. We experimentally compare BCO to imitation learning methods, including the state-of-the-art, generative adversarial imitation learning (GAIL) technique, and we show comparable task performance in several different simulation domains while exhibiting increased learning speed after expert trajectories become available.
hub tools
citation-role summary
background 2
method 1
citation-polarity summary
representative citing papers
ASH reaches 11.2/12 milestones in Pokemon Emerald and 9.9/12 in Zelda by self-improving via an IDM trained on its own trajectories to label internet video, while baselines plateau at roughly 6/12.
SeqRejectron constructs a stopping rule with a small set of validator policies to achieve horizon-free sample complexity for selective imitation learning under arbitrary dynamics shifts.
LEO enables efficient all-goals learning in goal-conditioned RL by jointly predicting for all goals in one network pass, yielding >250x speedup over relabelling and better performance on Craftax.
HITL-D combines diffusion policies with human input for shared robotic control, reducing required joystick axes and improving speed and workload in manipulation tasks per a 12-participant study.
TimeRewarder derives step-wise progress rewards from frame-wise temporal distances in passive videos and uses them to guide RL, achieving high success rates on Meta-World tasks with fewer interactions than prior methods or hand-designed rewards.
DAWM introduces a modular diffusion world model with an inverse dynamics model to produce complete synthetic transitions that improve conservative offline RL algorithms like TD3BC and IQL on D4RL tasks.
HybridVLA unifies diffusion and autoregression in a single VLA model via collaborative training and ensemble to raise robot manipulation success rates by 14% in simulation and 19% in real-world tasks.
Empirical analysis shows scaling inference compute via strategies like tree search can be more efficient than scaling model parameters, with 7B models plus novel search outperforming 34B models.
SemBid injects LLM-encoded Task, History, and Strategy semantics as tokens into offline bidding trajectories and uses self-attention to outperform numerical-only baselines in performance, constraint satisfaction, and robustness.
A framework decouples failure data for value estimation and success data for policy learning in offline RL to reduce collisions in robot navigation while maintaining success rates.
PTMT is a lightweight framework that automates parameter tuning for memory tiering via hybrid offline database building and online customized reinforcement learning, delivering 14-30% gains over defaults and 32% over prior art on four systems.
Veo-3 video predictions enable approximate task-level robot trajectories in zero-shot settings but require hierarchical integration with low-level VLA policies for reliable manipulation performance.
PACTS jointly model action trajectories and predicate belief trajectories in a single generative policy, enabling zero-shot skill composition via symbolic planning without retraining.
COOPO is a cyclic offline-online RL algorithm that repeatedly anchors the policy to a dataset via KL-regularized updates then fine-tunes online, claiming better sample efficiency and monotonic improvement under coverage assumptions.
UI-Oceanus shows that continual pre-training on forward dynamics predictions from synthetic GUI exploration improves agent success rates by 7% offline and 16.8% online, with gains scaling by data volume.
Robots outperform constrained human demonstrations by inferring state-only rewards from demos and using temporal interpolation to label and explore better trajectories, achieving 10x faster task completion on a real robotic arm than behavioral cloning.
LLMs cannot solve the medical treatment problem through imitation alone because it requires evidence from experiments or observations, posing ethical challenges for training such systems.
A centralized HRL planner with HTAN, multi-stage curricula, and counterfactual baseline scales multi-robot task planning to 200 robots and 1000 racks on unlearned maps in RMFS.
citing papers explorer
-
Demo-JEPA: Joint-Embedding Predictive Architecture for One-shot Cross-Embodiment Imitation
Demo-JEPA enables one-shot cross-embodiment imitation by mapping visual demonstrations to shared latent future trajectories that serve as subgoals for the target agent's own forward dynamics planning.
-
ASH: Agents that Self-Hone via Embodied Learning
ASH reaches 11.2/12 milestones in Pokemon Emerald and 9.9/12 in Zelda by self-improving via an IDM trained on its own trajectories to label internet video, while baselines plateau at roughly 6/12.
-
Learning When to Stop: Selective Imitation Learning Under Arbitrary Dynamics Shift
SeqRejectron constructs a stopping rule with a small set of validator policies to achieve horizon-free sample complexity for selective imitation learning under arbitrary dynamics shifts.
-
Goal-Conditioned Agents that Learn Everything All at Once
LEO enables efficient all-goals learning in goal-conditioned RL by jointly predicting for all goals in one network pass, yielding >250x speedup over relabelling and better performance on Craftax.
-
HITL-D: Human In The Loop Diffusion Assisted Shared Control
HITL-D combines diffusion policies with human input for shared robotic control, reducing required joystick axes and improving speed and workload in manipulation tasks per a 12-participant study.
-
TimeRewarder: Learning Dense Reward from Passive Videos via Frame-wise Temporal Distance
TimeRewarder derives step-wise progress rewards from frame-wise temporal distances in passive videos and uses them to guide RL, achieving high success rates on Meta-World tasks with fewer interactions than prior methods or hand-designed rewards.
-
DAWM: Diffusion Action World Models for Offline Reinforcement Learning via Action-Inferred Transitions
DAWM introduces a modular diffusion world model with an inverse dynamics model to produce complete synthetic transitions that improve conservative offline RL algorithms like TD3BC and IQL on D4RL tasks.
-
HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model
HybridVLA unifies diffusion and autoregression in a single VLA model via collaborative training and ensemble to raise robot manipulation success rates by 14% in simulation and 19% in real-world tasks.
-
Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Models
Empirical analysis shows scaling inference compute via strategies like tree search can be more efficient than scaling model parameters, with 7B models plus novel search outperforming 34B models.
-
On the Role of Language Representations in Auto-Bidding: Findings and Implications
SemBid injects LLM-encoded Task, History, and Strategy semantics as tokens into offline bidding trajectories and uses self-attention to outperform numerical-only baselines in performance, constraint satisfaction, and robustness.
-
Learning from Demonstration with Failure Awareness for Safe Robot Navigation
A framework decouples failure data for value estimation and success data for policy learning in offline RL to reduce collisions in robot navigation while maintaining success rates.
-
Hybrid Adaptive Tuning for Tiered Memory Systems
PTMT is a lightweight framework that automates parameter tuning for memory tiering via hybrid offline database building and online customized reinforcement learning, delivering 14-30% gains over defaults and 32% over prior art on four systems.
-
Veo-Act: How Far Can Frontier Video Models Advance Generalizable Robot Manipulation?
Veo-3 video predictions enable approximate task-level robot trajectories in zero-shot settings but require hierarchical integration with low-level VLA policies for reliable manipulation performance.
-
Jointly Learning Predicates and Actions Enables Zero-Shot Skill Composition
PACTS jointly model action trajectories and predicate belief trajectories in a single generative policy, enabling zero-shot skill composition via symbolic planning without retraining.
-
COOPO: Cyclic Offline-Online Policy Optimization Algorithm
COOPO is a cyclic offline-online RL algorithm that repeatedly anchors the policy to a dataset via KL-regularized updates then fine-tunes online, claiming better sample efficiency and monotonic improvement under coverage assumptions.
-
UI-Oceanus: Scaling GUI Agents with Synthetic Environmental Dynamics
UI-Oceanus shows that continual pre-training on forward dynamics predictions from synthetic GUI exploration improves agent success rates by 7% offline and 16.8% online, with gains scaling by data volume.
-
When a Robot is More Capable than a Human: Learning from Constrained Demonstrators
Robots outperform constrained human demonstrations by inferring state-only rewards from demos and using temporal interpolation to label and explore better trajectories, achieving 10x faster task completion on a real robotic arm than behavioral cloning.
-
Treatment, evidence, imitation, and chat
LLMs cannot solve the medical treatment problem through imitation alone because it requires evidence from experiments or observations, posing ethical challenges for training such systems.
-
Scalable Hierarchical Reinforcement Learning for Hyper Scale Multi-Robot Task Planning
A centralized HRL planner with HTAN, multi-stage curricula, and counterfactual baseline scales multi-robot task planning to 200 robots and 1000 racks on unlearned maps in RMFS.