# Digital Creatures Learn to Navigate in 3D | Two Minute Papers #153

## Метаданные

- **Канал:** Two Minute Papers
- **YouTube:** https://www.youtube.com/watch?v=2vnLBb18MuQ
- **Дата:** 14.05.2017
- **Длительность:** 2:45
- **Просмотры:** 12,689
- **Источник:** https://ekstraktznaniy.ru/video/14661

## Описание

The paper "DeepLoco: Dynamic Locomotion Skills
Using Hierarchical Deep Reinforcement Learning" is available here:
http://www.cs.ubc.ca/~van/papers/2017-TOG-deepLoco/index.html

Two Minute Papers Merch:
US: http://twominutepapers.com/
EU/Worldwide: https://shop.spreadshirt.net/TwoMinutePapers/

WE WOULD LIKE TO THANK OUR GENEROUS PATREON SUPPORTERS WHO MAKE TWO MINUTE PAPERS POSSIBLE:
Andrew Melnychuk, Christian Lawson, Dave Rushton-Smith, Dennis Abts, e, Esa Turkulainen, Michael Albrecht, Sunil Kim, VR Wizard.
https://www.patreon.com/TwoMinutePapers

Music: Antarctica by Audionautix is licensed under a Creative Commons Attribution license (https://creativecommons.org/licenses/by/4.0/)
Artist: http://audionautix.com/ 

Thumbnail background image credit: https://pixabay.com/photo-1505714/
Splash screen/thumbnail design: Felícia Fehér - http://felicia.hu

Károly Zsolnai-Fehér's links:
Facebook → https://www.facebook.com/TwoMinutePapers/
Twitter → https://twitter.com/karoly_zsolnai
Web → h

## Транскрипт

### <Untitled Chapter 1> []

dear fellow scholars this is two minute papers with károly on alpha here earlier we have talked about a few amazing

### Path Following [0:05]

algorithms to teach digital creatures to

### Walking on Narrow Paths [0:09]

walk and this time we are interested in controlling the joints of a digital character to not only work properly but take into consideration its surroundings this new version can navigate in 3d with

### Soccer Dribbling [0:20]

static and dynamic moving obstacles or even dribble a ball toward a target loving the execution and the production value of this paper this is accomplished

### Soccer Pirouette [0:31]

by an efficient system that consists of two controllers that are represented by learning algorithms one the low level controller is about maintaining balance

### Walking on Conveyor Belts [0:41]

and proper limb control by manipulating the joint positions and velocities appropriately this controller operates on a fine time scale 30 times per second and is trained via a four layer neural network to the high level controller can accomplish bigger overarching goals such as following a path or avoiding static

### Running on Conveyor Belts [1:02]

and dynamic obstacles we don't need to run this so often therefore to save resources this controller operates on a course time scale only twice each second

### Walking on Ice [1:13]

and is trained via a deep convolutional neural network it also has support for a small degree of transfer learning means that after successfully learning to solve a problem

### Running on Ice [1:23]

we don't have to start from scratch for the next one but we can reuse some of that valuable knowledge and get a head start this is a

### Pillars [1:31]

heavily researched area and is likely going to be one of the major next frontiers in machine learning research now make no mistake it is not like transfer learning is suddenly to be considered a soft problem but in this particular case it is finally a possibility really cool I hope this

### Blocks [1:47]

brief expose fired you up to this paper is a bomb make sure to have a look as always the link is available in the video description and by the way with

### Dynamic Obstacles [1:58]

your support on patreon we will soon be able to spend part of our budget on empowering research projects how amazing is that the new two-minute paper shirts are also flying off the shelves happy to hear you're enjoying them so much if you

### Narrow Paths (Width = 1m) [2:13]

are interested get up two minute papers calm if you are located in the US the EU and worldwide stores link is available in the video description thanks for watching and for your generous support and I'll see you next time