The context-aware learning model: Reward-based and experience-based logistic regression backpropagation
Joohee Suh, Dean F. Hougen · 2017
To deal with uncertain environments, autonomous agents need to be able to learn without supervision. However, reward-based interactive learning often exhibits limitations handling both generalizations and exceptions. For these reasons, this research introduces the Context-Aware Learning Model (CALM) and two different learning algorithms. CALM-rLRB combines logistic regression backpropagation in artificial neural networks with hyperbolic reward-based learning. CALM-eLRB adds an empirical knowledge base that enables experience-based learning. CALM is evaluated using four metrics on six synthetic data sets and shows promising performance.