no code implementations • 25 Jan 2021 • Keyan Zhai, Chu'an Li, Andre Rosendo
This paper presents the impact of a similar supports in the learning of a stable gait on a quadruped robot.
no code implementations • 4 Dec 2020 • Wangshu Zhu, Andre Rosendo
To address this issue we present a PPO variant, named Proximal Policy Optimization Smooth Algorithm (PPOS), and its critical improvement is the use of a functional clipping method instead of a flat clipping method.
no code implementations • 21 Jul 2020 • Jingyi Huang, Yizheng Zhang, Fabio Giardina, Andre Rosendo
While considering Sim and Real learning, our experiments show that the sample-efficient Deep Bayesian RL performance is better than DRL even when computation time (as opposed to number of iterations) is taken in consideration.
no code implementations • 8 Oct 2019 • Yizheng Zhang, Andre Rosendo
Deep Reinforcement Learning (DRL) has shown its promising capabilities to learn optimal policies directly from trial and error.