paper-with-me

홈 › Papers

Learning from the Hindsight Plan -- Episodic MPC Improvement

2016-09-28 · Aviv Tamar, Garrett Thomas, Tianhao Zhang, Sergey Levine, Pieter Abbeel

Model predictive control (MPC) is a popular control method that has proved effective for robotics, among other fields. MPC performs re-planning at every time step. Re-planning is done with a limited horizon per computational and real-time constraints and often also for robustness to potential model errors. However, the limited horizon leads to suboptimal performance. In this work, we consider the iterative learning setting, where the same task can be repeated several times, and propose a policy improvement scheme for MPC. The main idea is that between executions we can, offline, run MPC with a longer horizon, resulting in a hindsight plan. To bring the next real-world execution closer to the hindsight plan, our approach learns to re-shape the original cost function with the goal of satisfying the following property: short horizon planning (as realistic during real executions) with respect to the shaped cost should result in mimicking the hindsight plan. This effectively consolidates long-term reasoning into the short-horizon planning. We empirically evaluate our approach in contact-rich manipulation tasks both in simulated and real environments, such as peg insertion by a real PR2 robot.

📄 PDF Abstract BibTeX arXiv:1609.09001

Code (1)

zuoxingdong/VIN_PyTorch_Visdom pytorch

Tasks

Contact-rich ManipulationModel Predictive Control

Similar Papers 제목 키워드 기반

Episodic Self-Imitation Learning with Hindsight

2020-11-26 · Tianhong Dai, Hengyan Liu, Anil Anthony Bharath

Episodic self-imitation learning, a novel self-imitation algorithm with a trajectory selection module and an adaptive loss function, is proposed to speed up reinforcement learning. Compared to the original self-imitation…

continuous-controlContinuous ControlImitation Learning

Faster Reinforcement Learning with Value Target Lower Bounding

2021-09-29 · Le Zhao, Wei Xu

We show that an arbitrary lower bound of the optimal value function can be used to improve the Bellman value target during value learning. In the tabular case, value learning under the lower bounded Bellman operator con…

Atari Gamescontinuous-controlContinuous Controlreinforcement-learning+2

Translating Flow to Policy via Hindsight Online Imitation

2025-12-22 · Yitian Zheng, Zhangchen Ye, Weijun Dong, Shengjie Wang 외 arxiv

Recent advances in hierarchical robot systems leverage a high-level planner to propose task plans and a low-level policy to generate robot actions. This design allows training the planner on action-free or even non-robot…

Generalizable Episodic Memory for Deep Reinforcement Learning

2021-03-11 · Hao Hu, Jianing Ye, Guangxiang Zhu, Zhizhou Ren 외

Episodic memory-based methods can rapidly latch onto past successful strategies by a non-parametric memory and improve sample efficiency of traditional reinforcement learning. However, little effort is put into the conti…

Atari Gamescontinuous-controlContinuous ControlDeep Reinforcement Learning+4

Using Hindsight to Anchor Past Knowledge in Continual Learning

2020-02-19 · Arslan Chaudhry, Albert Gordo, Puneet K. Dokania, Philip Torr 외

In continual learning, the learner faces a stream of data whose distribution changes over time. Modern neural networks are known to suffer under this setting, as they quickly forget previously acquired knowledge. To addr…

Bilevel OptimizationContinual Learning