Learning to Walk via Deep Reinforcement Learning. Tuomas Haarnoja, Sehoon Ha, Aurick Zhou, Jie Tan, George Tucker, Sergey Levine. Robotics: Science and Systems (RSS). 2019.
Abstract
Deep reinforcement learning (deep RL) holds the promise of automating the acquisition of complex controllers that can map sensory inputs directly to low-level actions. In the domain of robotic locomotion, deep RL could enable learning locomotion skills with minimal engineering and without an explicit model of the robot dynamics. Unfortunately, applying deep RL to real-world robotic tasks is exceptionally difficult, primarily due to poor sample complexity and sensitivity to hyperparameters. While hyperparameters can be easily tuned in simulated domains, tuning may be prohibitively expensive on physical systems, such as legged robots, that can be damaged through extensive trial-and-error learning. In this paper, we propose a sample-efficient deep RL algorithm based on maximum entropy RL that requires minimal per-task tuning and only a modest number of trials to learn neural network policies. We apply this method to learning walking gaits on a real-world Minitaur robot. Our method can acquire a stable gait from scratch directly in the real world in about two hours, without relying on any model or simulation, and the resulting policy is robust to moderate variations in the environment. We further show that our algorithm achieves state-of-the-art performance on simulated benchmarks with a single set of hyperparameters. Videos of training and the learned policy can be found on the project website.
1 view
1761
480
2 months ago 00:05:24 1
Me or We? Cultural Difference between East and West
2 months ago 00:20:21 1
ONLY THE CHOSEN ONES CARRY THIS SIGN
2 months ago 00:01:00 1
в этой стране мусор исчезает за один день, а из пластиковых бутылок делают лучшие дороги
2 months ago 01:04:12 1
Agatha All Along | ASMR | 1hr
2 months ago 00:02:20 1
Decadent | “Madness” Trailer
2 months ago 00:00:47 1
Reveal Trailer | My Time at Evershine
2 months ago 00:31:44 1
How to draw a mountain sunset landscape / Learn acrylic painting / Art painting.
2 months ago 00:22:10 1
Ex Witch Exposes the Danger of Speaking Like This | The Power of Silence (Erica Mukisa)
2 months ago 00:05:01 2
xNurbs in Plasticity - IT JUST WORKS ?!
2 months ago 00:05:08 1
ALBINO THREADFIN ACARA a *CENTERPIECE FISH!*
2 months ago 00:03:56 1
Bells Hells Campaign 3 Opening Title 2023 | It’s Thursday Night (Critical Role Theme)
2 months ago 00:37:50 1
How to draw running horse with pencil step by step.
2 months ago 00:20:47 1
[ENG] FIFTY FIFTY balances out their aegyo & charisma ⚖️ ㅣ Spotify Rookie-Thru
2 months ago 01:07:26 1
A One World Religion is Secretly Taking Over American Churches w/ James Lindsay
2 months ago 00:24:01 1
Globetrotter: Mastering i18n for User-Centric Web Applications | VueConf US 2024
2 months ago 00:00:58 1
Discover Pullman Resort Al Marjan Island!🏤🏖 | Air Way To Go 🌍✈
2 months ago 00:21:50 1
One Reason Modern Entertainment is Failing
2 months ago 00:27:13 1
طلاء أكريليك للنهر وا
2 months ago 00:02:53 1
ПРИНЯЛИ ПОЛИЦЕЙСКИЕ | СТАНТ НА СКУТЕРЕ | СТАНТ НА ПИТБАЙКЕ
2 months ago 00:19:32 1
Painting a Realistic River With Acrylic. Time Lapse /90
2 months ago 00:03:13 1
Are You Living an Insta Lie? Social Media Vs. Reality
2 months ago 01:31:18 1
50: Paid In Respect
2 months ago 00:02:25 1
The Umbrella Academy: Season 4 | Bloopers | Netflix
2 months ago 00:00:25 1
How to use AI for automation #mindosol #aitools #freeaitools #mindosolai