Asynchronous Methods for Deep Reinforcement Learning: TORCS

Channel:
United Kingdom Google DeepMind
Subscribers:
663,000
Published on ● Video Link: https://www.youtube.com/watch?v=0xo1Ldx3L5Q


Duration: 0:30
46,998 views
335

The video shows an agent driving a racecar using only raw pixels as input. The agent was trained using the Asynchronous Advantage Actor-Critic (A3C) algorithm. During training, the agent was rewarded for maintaining high velocity along the center of the racetrack.
Paper link - http://arxiv.org/pdf/1602.01783.pdf



Other Videos By Google DeepMind


2016-12-05DeepMind Lab - Nav Maze Level 1
2016-11-22DeepMind Health – Michael Wise – a patient's journey
2016-11-22Streams – a platform for a digital NHS
2016-11-22DeepMind Health – Partnership with the Royal Free London NHS Foundation Trust
2016-11-04StarCraft II DeepMind feature layer API
2016-10-12Differentiable neural computer family tree inference task
2016-09-20DeepMind Health - Patient and Public Engagement Event
2016-09-08DeepMind Health at NHS Expo 2016 – Delivering the Benefits of a Digital NHS
2016-07-05Mustafa Suleyman – New Ways for Technology to Enhance Patient Care – King's Fund 5th July 2016
2016-07-05DeepMind Health – Moorfields Eye Hospital London Collaboration
2016-06-14Asynchronous Methods for Deep Reinforcement Learning: TORCS
2016-06-14Asynchronous Methods for Deep Reinforcement Learning: MuJoCo
2016-06-14Asynchronous Methods for Deep Reinforcement Learning: Labyrinth
2016-06-13DQN Breakout
2016-06-13DQN SPACE INVADERS
2016-03-18Match 5 15 Minute Summary - Google DeepMind Challenge Match 2016
2016-03-15Match 5 - Google DeepMind Challenge Match: Lee Sedol vs AlphaGo
2016-03-13Match 4 15 Minute Summary - Google DeepMind Challenge Match 2016
2016-03-13Match 4 - Google DeepMind Challenge Match: Lee Sedol vs AlphaGo
2016-03-12Match 3 15 Minute Summary - Google DeepMind Challenge Match 2016
2016-03-12Match 2 15 Minute Summary - Google DeepMind Challenge Match 2016