A paper in Applied AI Letters argues that single-objective RL defenders built on hand-weighted rewards cannot adapt at inference time to competing goals such as stopping intrusions versus avoiding downtime. It presents a simple multi-objective network defence game in which defending against red agents must be balanced with preserving network services.
Why it matters
It formalises the trade-off between defensive action and operational disruption that any autonomous defender must manage.
Key facts
As stated in the sources, with where to find them.
- The authors contrast single-objective RL with a handcrafted weighted reward against a multi-objective formulation that keeps competing objectives separate.Abstract
Findings that cite this record
No tracked finding cites this record yet.
Sources
Related records
Aug 30, 2025
Aug 8, 2025
Jan 8, 2026
Jun 27, 2025
May 7, 2025
Aug 11, 2024