US 12,468,934 B2
Systems and methods for generating dynamic conversational responses using deep conditional learning
Minh Le, McLean, VA (US); Rui Zhang, McLean, VA (US); and Jonathan Gross, McLean, VA (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on May 11, 2021, as Appl. No. 17/316,998.
Prior Publication US 2022/0366233 A1, Nov. 17, 2022
Int. Cl. G06N 3/08 (2023.01); G06F 16/9032 (2019.01); G06F 18/20 (2023.01); G06F 40/30 (2020.01); G06N 3/045 (2023.01); G06N 5/022 (2023.01)
CPC G06N 3/08 (2013.01) [G06F 16/90332 (2019.01); G06F 18/295 (2023.01); G06F 40/30 (2020.01); G06N 3/045 (2023.01); G06N 5/022 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system for generating dynamic conversational responses using conditional deep learning, the system comprising:
cloud-based storage circuitry configured to store a first neural network that is trained based on a first training regime and a second training regime, wherein the first training regime comprises training an initial version of the first neural network based on a first data set and using knowledge distillation to mimic predictions made by a first model by minimizing a loss function in which a target is a distribution of class probabilities predicted by the first model, wherein the second training regime comprises further training the initial version based on a second data set and using a second neural network as a constraint, wherein the second neural network (i) is trained based on a third data set, (ii) has parameters that remain unchanged during the second training regime, (iii) comprises an output layer for the initial version during the second training regime, and (iv) weights of the second neural network are updated by switching labels of the initial version at a minibatch level, and wherein first data set comprises data on a plurality of users, and wherein the second data set comprises feedback data on actual intents of a user;
cloud-based control circuitry configured to:
receive user data in response to a user interacting with a user interface;
generate a feature input based on the user data;
input the feature input into a first neural network, wherein the first neural network is trained based on a first training regime and a second training regime;
receive an output from the first neural network; and
cloud-based input/output circuitry configured to:
generate for simultaneous display, on the user interface, a first dynamic conversational response and a second dynamic conversational response based on the output, wherein the first dynamic conversational response corresponds to a first probable intent of the user, wherein the second dynamic conversational response corresponds to a second probable intent of the user, and wherein each dynamic conversational response is generated for display with its respective probability.