Journal of Machine Learning Research · 2017 · 96 citations · 0 references
ComputingComputer ScienceArtificial IntelligenceMachine LearningOptimizationDecision-making ModelsExploration-exploitation TradeoffPolicy LearningPreference ModelingReinforcement LearningReinforcement Learning FoundationsExploration - Exploitation Trade-offExploration-exploitation Trade-offPreference LearningPreference-based LearningPreference-based Reward LearningReinforcement Learning From Human FeedbackReward Function DesignReward Shaping
Reinforcement learning (RL) techniques optimize the accumulated long-term reward of a suitably chosen reward function. However, designing such a reward function often requires a lot of task-specifi...