On April 18, in Recentive Analytics, Inc., v. Fox Corp., which presented a question of first impression, the Federal Circuit held that claims that do no more than apply established methods of machine learning to a new data environment are not patent eligible under 35 U.S.C. § 101.
The case involved four patents owned by Recentive that purport to solve problems confronting the entertainment industry and television broadcasters. Such problems include how to optimize the scheduling of live events (e.g., NFL games) and how to optimize “network maps,” which determine the programs or content displayed by a broadcaster’s channels within certain geographic markets at particular times. The concepts of preparing network maps and schedules for live events have existed for a long time with the tasks having previously been performed manually by humans. The patents claim the use of machine learning for the generation of network maps and schedules for television broadcasts and live events.
Recentive sued Fox, alleging infringement of the four patents. The district court granted Fox’s motion to dismiss, concluding that the patents were ineligible under § 101 under the two-step inquiry of Alice Corporation v. CLS Bank International.
On appeal, the Federal Circuit reviewed the district court’s determination of patent ineligibility under § 101 and affirmed. Specifically, under the first step of the Alice inquiry, the Federal Circuit found that the disputed claims are clearly directed to ineligible, abstract subject matter. The court reasoned that Recentive repeatedly conceded it was not claiming machine learning itself, and that the patents rely on the use of generic machine learning technology in carrying out the claimed methods for generating event schedules and network maps. The court supported this reasoning by noting that the machine learning technology described in the patents is conventional, with the patents stating that any suitable machine learning technique can be used.
Recentive argued its application of machine learning is not generic because the algorithms function dynamically, so the maps and schedules are automatically customizable and updated with real-time data. But the court responded to these arguments by noting that Recentive admitted the patents do not claim a specific method for improving the mathematical algorithm or making machine learning better, and that the claims do not delineate steps through which the machine learning technology achieves an improvement.
The court also noted that the requirement that the machine learning model be iteratively trained or dynamically adjusted as described in some of Recentive’s patents does not represent a technological improvement. This is because Recentive’s own representations about the nature of machine learning vitiated this argument by indicating, for example, that using a machine learning technique necessarily includes an iterative training step.
The court went on to find that the only thing the disputed claims disclose about the use of machine learning is that machine learning is used in a new environment, which is event scheduling and the creation of network maps. Recentive acknowledged that before the introduction of machine learning, event schedules and network maps were generated manually by humans. The court saw no merit to Recentive’s argument that its patents are eligible because they apply machine learning to this new field of use. The court cited its long-recognized view that an abstract idea does not become nonabstract by limiting the invention to a particular field of use or technological environment.
At Alice step two, the court found nothing in the claims that would transform the patents into something significantly more than the abstract idea of generating event schedules and network maps through the application of machine learning. Recentive had argued that the inventive concept is using machine learning to dynamically generate optimized maps and schedules based on real-time data and update them based on changing conditions. But the Federal Circuit agreed with the district court that this argument is no more than claiming the abstract idea itself.
Therefore, the Federal Circuit affirmed the district court’s decision and held that patents that do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning models to be applied, are patent ineligible under § 101.
For more information on this decision, please contact Fitch Even partner Richard E. Wawrzyniak, author of this alert.
Fitch Even IP Alert®
Richard E. Wawrzyniak
Richard E. Wawrzyniak’s practice focus is on patent preparation, prosecution, and strategy for high-technology companies, assisting clients in the development, protection, and management of intellectual property rights in the electrical, software, Internet/e-commerce, and mechanical arts.
