arXiv subjects

Ted Xiao

Publications and source records attributed to Ted Xiao.

39 records · Page 3Linked to original sources

arXiv2020

Thinking While Moving: Deep Reinforcement Learning with Concurrent Control

We study reinforcement learning in settings where sampling an action from the policy must be done concurrently with the time evolution of the controlled system, such as when a robot must decide on the next action while still performing the previous action. Much like a person or an animal, the robot must think and move at the same time, deciding on its next action before the previous one has completed. In order to develop an algorithmic framework for such concurrent control problems, we start with a continuous-time formulation of the Bellman equations, and then discretize them in a way that is aware of system delays. We instantiate this new class of approximate dynamic programming methods via a simple architectural extension to existing value-based deep reinforcement learning algorithms. We evaluate our methods on simulated benchmark tasks and a large-scale robotic grasping task where the robot must "think while moving".

cs.LG↗

arXiv2019

Learning Latent Plans from Play

Acquiring a diverse repertoire of general-purpose skills remains an open challenge for robotics. In this work, we propose self-supervising control on top of human teleoperated play data as a way to scale up skill learning. Play has two properties that make it attractive compared to conventional task demonstrations. Play is cheap, as it can be collected in large quantities quickly without task segmenting, labeling, or resetting to an initial state. Play is naturally rich, covering ~4x more interaction space than task demonstrations for the same amount of collection time. To learn control from play, we introduce Play-LMP, a self-supervised method that learns to organize play behaviors in a latent space, then reuse them at test time to achieve specific goals. Combining self-supervised control with a diverse play dataset shifts the focus of skill learning from a narrow and discrete set of tasks to the full continuum of behaviors available in an environment. We find that this combination generalizes well empirically---after self-supervising on unlabeled play, our method substantially outperforms individual expert-trained policies on 18 difficult user-specified visual manipulation tasks in a simulated robotic tabletop environment. We additionally find that play-supervised models, unlike their expert-trained counterparts, are more robust to perturbations and exhibit retrying-till-success behaviors. Finally, we find that our agent organizes its latent plan space around functional tasks, despite never being trained with task labels. Videos, code and data are available at learning-from-play.github.io

cs.RO↗

arXiv2017

Goal-Driven Dynamics Learning via Bayesian Optimization

Real-world robots are becoming increasingly complex and commonly act in poorly understood environments where it is extremely challenging to model or learn their true dynamics. Therefore, it might be desirable to take a task-specific approach, wherein the focus is on explicitly learning the dynamics model which achieves the best control performance for the task at hand, rather than learning the true dynamics. In this work, we use Bayesian optimization in an active learning framework where a locally linear dynamics model is learned with the intent of maximizing the control performance, and used in conjunction with optimal control schemes to efficiently design a controller for a given task. This model is updated directly based on the performance observed in experiments on the physical system in an iterative manner until a desired performance is achieved. We demonstrate the efficacy of the proposed approach through simulations and real experiments on a quadrotor testbed.

eess.SY↗