Improving Neural Network Learning Through Dual Variable Learning Rates

This paper introduces and evaluates a novel training method for neural networks: Dual Variable Learning Rates (DVLR). Building on insights from behavioral psychology, the dual learning rates are used to emphasize correct and incorrect responses differently, thereby making the feedback to the network more specific. Further, the learning rates are varied as a function of the network's performance, thereby making it more efficient. DVLR was implemented on three types of networks: feedforward, convolutional, and residual, and two domains: MNIST and CIFAR-10. The results suggest a consistently improved accuracy, demonstrating that DVLR is a promising, psychologically motivated technique for training neural network models.

[1]  Choong Woong Lee,et al.  Fast competitive learning with classified learning rates for vector quantization , 1995, Signal Process. Image Commun..

[2]  Leslie N. Smith,et al.  Cyclical Learning Rates for Training Neural Networks , 2015, 2017 IEEE Winter Conference on Applications of Computer Vision (WACV).

[3]  C. B. Ferster,et al.  Schedules of reinforcement , 1957 .

[4]  Floyd C. Mace,et al.  Schedules of reinforcement , 2011 .

[5]  Bruno B. Averbeck,et al.  Amygdala and ventral striatum population codes implement multiple learning rates for reinforcement learning , 2017, 2017 IEEE Symposium Series on Computational Intelligence (SSCI).

[6]  W. Brown Animal Intelligence: Experimental Studies , 1912, Nature.