This text explores USPTO dealing with of Alice points involving synthetic intelligence and machine studying by a sampling of latest Patent Trial and Enchantment Board (PTAB) selections. See Alice Corp. v. CLS Int’l, 134 S. Ct. 2347 (2014). Some selections dutifully utilized USPTO pointers on material eligibility, together with Instance 39 thereof, to resolve enchantment points dropped at the PTAB. In a single case, the PTAB sua sponte provided eligibility steerage even with no Alice enchantment subject earlier than it. These selections inform methods to optimize patent drafting and prosecution for synthetic intelligence and machine studying associated innovations.
“Generic Machine Studying Algorithm”
In Ex parte Hussain, Enchantment No. 2020-005406 (PTAB Feb. 18, 2021), the PTAB thought of the subject material eligibility of claims reciting a “machine studying algorithm” in relation to mitigation of threat of shopper default on a web-based transaction. Consultant declare 1 recited as follows:
- A pc-implemented technique, comprising: beneath the management of a number of pc programs that execute directions,offering executable directions to a consumer computing gadget related to a consumer that, on account of being executed by the consumer computing gadget, causes the consumer computing gadget to:
accumulate consumer information that features:
personally identifiable details about the consumer; an identifier related to the consumer computing gadget; and a measurement captured by the consumer computing gadget related the consumer interacting with the consumer computing gadget, the measurement together with:
an motion carried out by the consumer to an object displayed in a consumer interface of the consumer computing gadget; an id of the article; and a time worth indicating a time at which the motion was carried out to the article; and supply the consumer information to the a number of pc programs;
acquiring saved transactional information related to a number of earlier transactions involving the consumer;
acquiring verification information verifying that the personally identifiable data is correct;
reworking the saved transactional information, the verification information, and the consumer information that features the measurement right into a set of variable values usable as enter right into a machine studying algorithm that’s skilled to deduce traits concerning the consumer from the set of variable values;
acquiring, on account of inputting the set of variable values into the machine studying algorithm, a constancy rating output by the machine studying algorithm; and
primarily based at the least partly on the constancy rating and with out acquiring extra details about the consumer from a 3rd occasion:
figuring out a cost or credit score choice to show within the consumer interface for a present transaction; and updating, contents of the consumer interface to supply the cost or credit score choice.