A Penalty Default Approach to Preemptive Harm Disclosure and Mitigation for AI Systems.
Rui-Jie YewDylan Hadfield-MenellPublished in: AIES (2022)
Keyphrases
- ai systems
- intelligent systems
- artificial intelligence
- scheduling problem
- objective function
- information loss
- natural language generation
- lower bound
- risk management
- single machine
- statistical databases
- scheduling algorithm
- processing times
- game play
- privacy protection
- computational intelligence
- penalty function
- social networks
- np hard
- data quality
- neural network
- recommender systems
- case study
- privacy concerns
- e learning
- priority queue