Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions

no code implementations • 28 Mar 2022 • Alejandro Escontrela, Xue Bin Peng, Wenhao Yu, Tingnan Zhang, Atil Iscen, Ken Goldberg, Pieter Abbeel

We also demonstrate that an effective style reward can be learned from a few seconds of motion capture data gathered from a German Shepherd and leads to energy-efficient locomotion strategies with natural gait transitions.

Disentangled Planning and Control in Vision Based Robotics via Reward Machines

no code implementations • 28 Dec 2020 • Alberto Camacho, Jacob Varley, Deepali Jain, Atil Iscen, Dmitry Kalashnikov

In this work we augment a Deep Q-Learning agent with a Reward Machine (DQRM) to increase speed of learning vision-based policies for robot tasks, and overcome some of the limitations of DQN that prevent it from converging to good-quality policies.

Q-Learning

From Pixels to Legs: Hierarchical Learning of Quadruped Locomotion

no code implementations • 23 Nov 2020 • Deepali Jain, Atil Iscen, Ken Caluwaerts

We show that hierarchical policies can concurrently learn to locomote and navigate in these environments, and show they are more efficient than non-hierarchical neural network policies.

Hierarchical Reinforcement Learning Navigate

Policies Modulating Trajectory Generators

3 code implementations • 7 Oct 2019 • Atil Iscen, Ken Caluwaerts, Jie Tan, Tingnan Zhang, Erwin Coumans, Vikas Sindhwani, Vincent Vanhoucke

We propose an architecture for learning complex controllable behaviors by having simple Policies Modulate Trajectory Generators (PMTG), a powerful combination that can provide both memory and prior knowledge to the controller.

Data Efficient Reinforcement Learning for Legged Robots

no code implementations • 8 Jul 2019 • Yuxiang Yang, Ken Caluwaerts, Atil Iscen, Tingnan Zhang, Jie Tan, Vikas Sindhwani

We present a model-based framework for robot locomotion that achieves walking based on only 4. 5 minutes (45, 000 control steps) of data collected on a quadruped robot.

reinforcement-learning Reinforcement Learning (RL) +1

Hierarchical Reinforcement Learning for Quadruped Locomotion

no code implementations • 22 May 2019 • Deepali Jain, Atil Iscen, Ken Caluwaerts

We test our framework on a path-following task for a dynamic quadruped robot and we show that steering behaviors automatically emerge in the latent command space as low-level skills are needed for this task.

Hierarchical Reinforcement Learning reinforcement-learning +1

Provably Robust Blackbox Optimization for Reinforcement Learning

no code implementations • 7 Mar 2019 • Krzysztof Choromanski, Aldo Pacchiano, Jack Parker-Holder, Yunhao Tang, Deepali Jain, Yuxiang Yang, Atil Iscen, Jasmine Hsu, Vikas Sindhwani

Interest in derivative-free optimization (DFO) and "evolutionary strategies" (ES) has recently surged in the Reinforcement Learning (RL) community, with growing evidence that they can match state of the art methods for policy optimization problems in Robotics.

reinforcement-learning Reinforcement Learning (RL) +1

NoRML: No-Reward Meta Learning

1 code implementation • 4 Mar 2019 • Yuxiang Yang, Ken Caluwaerts, Atil Iscen, Jie Tan, Chelsea Finn

To this end, we introduce a method that allows for self-adaptation of learned policies: No-Reward Meta Learning (NoRML).

Meta-Learning Reinforcement Learning (RL)

Sim-to-Real: Learning Agile Locomotion For Quadruped Robots

no code implementations • 27 Apr 2018 • Jie Tan, Tingnan Zhang, Erwin Coumans, Atil Iscen, Yunfei Bai, Danijar Hafner, Steven Bohez, Vincent Vanhoucke

The control policies are learned in a physics simulator and then deployed on real robots.