Computer Science
a year ago

Detecting Novelty in Open-World Environments: Insights from Angry Birds Game.

2 views
arXiv (Cornell University)
The Difficulty of Novelty Detection in Open-World Physical Domains: An Application to Angry Birds
Vimukthini Pinto, Cheng Xue, Chathura Gamage, Matthew Stephenson, Jochen Renz

Paper Summary

Detecting new things in open-world environments is important for both humans and AI. Researchers created a way to measure how hard it is to spot new things in a physics-based world. They tested this method in Angry Birds and found that it matched how people detect new things.

Related papers

Theme background image
Computer Science
a year ago

AI struggles to adapt in new environments, impacting real-world applications.

Anomaly Detection
Empirical Studies
Malware Detection
2 views
Theme background image
Computer Science
2 years ago

New library detects anomalies and novelties in data streams effortlessly!

Anomaly Detection
Intrusion Detection
Concept Drift
3 views
Theme background image
Computer Science
4 years ago

Unveiling hidden data: Novelty detection revolutionizes machine learning applications.

Anomaly Detection
Intrusion Detection
Digital Epidemiology
1 view