Maintaining the Illusion of Reality: Transfer in RL by Keeping Agents in the DARC – Machine Learning Blog | ML@CMU | Carnegie Mellon University

Maintaining the Illusion of Reality:  Transfer in RL by Keeping Agents in the DARC – Machine Learning Blog | ML@CMU | Carnegie Mellon University

Reinforcement learning (RL) is often touted as a promising approach for costly and risk-sensitive applications, yet practicing and learning in those domains directly is expensive. It costs time (e.g., OpenAI's Dota2 project used 10,000 years of experience), it costs money (e.g., "inexpensive" roboti

3 mentions: @rsalakhu@mlcmublog@asifrazzaq1988
Date: 2020/07/31 15:01

Referring Tweets

@mlcmublog How can a robot use practice in one area to solve tasks in another? Ben Eysenbach explains one effective approach: teach robots to avoid actions that indicate whether they're in the practice area or the real world. Post: t.co/lejAUWlwGK Paper: t.co/atW1HIbCBM

Related Entries

Read more GitHub - onnx/tutorials: Tutorials for creating and using ONNX models
0 users, 2 mentions 2020/05/02 03:51
Read more Learning to Explore using Active Neural SLAM – Machine Learning Blog | ML@CMU | Carnegie Mellon Univ...
0 users, 2 mentions 2020/06/19 15:00
Read more Carnegie Mellon University at ICML 2020 – Machine Learning Blog | ML@CMU | Carnegie Mellon Universit...
0 users, 5 mentions 2020/07/13 12:47
Read more In a World Where… AI is an Everyday Part of Business - Open Source Leader in AI and ML
0 users, 3 mentions 2020/07/22 02:30
Read more Exploring the Next Frontier of Automatic Machine Learning with H2O Driverless AI - Open Source Leade...
0 users, 2 mentions 2020/07/28 22:28