A Threshold-based Scheme for Reinforcement Learning in Neural Networks
This proposes a new learning paradigm for neural networks, potentially impacting researchers seeking alternatives to backpropagation.
The paper presents a threshold-based reinforcement learning scheme for neural networks that can solve linearly inseparable problems and form long-term strategies, offering a biologically inspired alternative to backpropagation for both supervised and unsupervised training.
A generic and scalable Reinforcement Learning scheme for Artificial Neural Networks is presented, providing a general purpose learning machine. By reference to a node threshold three features are described 1) A mechanism for Primary Reinforcement, capable of solving linearly inseparable problems 2) The learning scheme is extended to include a mechanism for Conditioned Reinforcement, capable of forming long term strategy 3) The learning scheme is modified to use a threshold-based deep learning algorithm, providing a robust and biologically inspired alternative to backpropagation. The model may be used for supervised as well as unsupervised training regimes.