Reinforcement Learning Patents: A Transatlantic Review

Abstract

One of the most difficult problems with developing scalable artificial intelligence (AI) is tracking technical innovation. As such, this research will provide empirical data relating to patents with legal claims to state of the art in AI technologies, reinforcement learning. A keystone architecture in the machine learning paradigm, reinforcement learning technologies power trading algorithms, driverless cars, and space satellites. In competitive global markets, owning and protecting legal rights to reinforcement learning technologies may mean the difference between market dominance and irrelevance for the transatlantic firm. Aggregating patent data for reinforcement learning technologies from both the United States Patent and Trademark Office (USPTO) and the European Patent Office (EPO), this research offers a transatlantic perspective on reinforcement learning innovation and intellectual property protections.