PofoliaShared via Pofolia

arXiv (Cornell University)· 2017· Preprint

Diagnosing Non-Intermittent Anomalies in Reinforcement Learning Policy Executions (Short Paper)

Schulman, John, Wolski, Filip, Dhariwal, Prafulla, Alec Radford et al.

Short summary

A sim-to-real transfer of an octocopter RL controller revealed a 100% increase in real-world flight deviation, though still within 2m safety corridors, and significantly different vehicle orientation due to a real-world turning flight mode.

AI-generated from the title and abstract; the full text is not read.

TakeawaysIn the app
Key pointsIn the app
Ask the paperIn the app

The rest is in the Pofolia app

Takeaways, key points and questions to the paper; new summaries every day for your field. Free.

Sign in on the web to open

Field: Artificial Intelligence

Artificial IntelligenceComputer Science