Reinforcement learning for discounted values often loses the goal in the application to animal learning.
Yamaguchi Y, Sakai Y
Neural networks
35
88-91