Download This PaperDeep Reinforcement Learning Patents: An Empirical Survey
Brian Haney, Deep Reinforcement Learning Patents: An Empirical Survey, UCLA J. L. & Tech __ (2021).
68 Pages Posted: 20 Apr 2020 Last revised: 4 Feb 2021
Date Written: April 6, 2020
Abstract
Deep reinforcement learning is a new type of machine learning resulting from the technical convergence of two more mature machine learning methods, deep learning and reinforcement learning. Deep reinforcement learning is important because it is a scalable method for general intelligence – a machine capable of achieving any definable goal.
Since 1995, the United States Patent and Trademark Office has granted over 600 patents with claim terms relating to deep reinforcement learning. Yet, while the literature on software patents is visibly scaling – the literature specifically focused on deep reinforcement learning patents is non-existent. First, this Article discusses technical approaches to deep reinforcement learning; second it contributes the first empirical patent review for deep reinforcement learning technologies, including market modeling, legal claims analysis, and valuation strategies.
Keywords: Deep Learning, Deep Reinforcement Learning, Patents, Patent Valuation, Artificial Intelligence
Suggested Citation: Suggested Citation