Human-aligned artificial intelligence is a multiobjective problem.
Peter VamplewRichard DazeleyCameron FoaleSally FirminJane MummeryPublished in: Ethics Inf. Technol. (2018)
Keyphrases
- multi objective
- artificial intelligence
- artificially intelligent
- evolutionary algorithm
- cognitive psychology
- optimization algorithm
- science fiction
- expert systems
- conflicting objectives
- computational intelligence
- human intelligence
- multiobjective optimization
- multiple objectives
- human reasoning
- human level
- genetic algorithm
- ai systems
- neural network
- software engineering
- intelligent agents
- intelligent systems
- artificial neural networks
- human learning
- pareto optimal
- knowledge based systems
- multiobjective evolutionary algorithms
- multi objective optimization
- human subjects
- objective function
- human experts
- multiobjective genetic algorithm
- human problem solving
- intelligent machines
- optimum design
- computer science
- ai methods
- simulated annealing
- knowledge representation
- fuzzy logic
- machine intelligence
- human beings
- nsga ii
- human behavior
- cognitive science
- human activities
- evolutionary computation