Capital One Services, LLC

International Business Machines Corporation

Capital Services, LLC

Erik Thomas Mueller

Oluwatobi Olabiyi

Scott Karp

Zachary Kulis

Rui Zhang

Omar Florez Choque

Alexandra Coman

Christopher Larson

David Angelo Ferrucci

Anthony Tom Levas

Sugato Bagchi

Minh Le

Victor Alvarez Miranda

Anish Khazane

Gregory W Zoller

Sujay Eliphaz Jacob

Stephanie Hay

Adam Roy Paynter

Vinay Igure

Kunlaya Soiaporn

Diana Mingels

Varun Singh

Tanushree Luke

Alan Salimov

Margaret Mayer

Juhnyoung Lee

Rangachari Anand

Moninder Singh

Leora Morgenstern

Tarek Aziz Lahlou

Leiguang Gong

Martin S Kohn

Liwei Dai

Doug Riecken

Melvin Perez-Cedano

Raymond R Hitney

Jennifer Chu-Carroll

Adam P Lally

Sara Mikulic

William Miller

Embodiments disclosed herein generally relate to a system and method for proactively generating an intervening message for a remote client device in response to an anticipated user action. A computing system receives one or more streams of user activity. The one or more streams of user activity include interaction with a server of an organization via an application executing on the remote client device. The computing system inputs the one or more streams of user activity into a prediction model. The computing system identifies an anticipated user action based on a prediction output from the prediction model. The computing system determines, based on a solution model, a proposed solution to the anticipated user action. The computing system generates an anticipated message to be transmitted to the remote client device of the user. The computing system transmits the anticipated message to the remote client device of the user.

System and method for proactive intervention to reduce high cost channel usage

A system for using hash keys to preserve privacy across multiple tasks is disclosed. The system may provide training batch(es) of input observations each having a customer request and stored task to an encoder, and assign a hash key(s) to each of the stored tasks. The system may provide a new batch of input observations with a new customer request and new task to the encoder. The encoder may generate a new hash key assigned to the new customer request and determine whether any existing hash key corresponds with the new hash key. If so, the system may associate the new batch of input observations with the corresponding hash key and update the corresponding hash key such that it is also configured to provide access to the new batch of input observations. If not, the system may generate a new stored task and assign the new hash key to it.

Systems using hash keys to preserve privacy across multiple tasks

Machine classifiers in accordance with embodiments of the invention capture long-term temporal dependencies in the dialogue data better than the existing recurrent neural network-based architectures. Additionally, machine classifiers may model the joint distribution of the context and response as opposed to the conditional distribution of the response given the context as employed in sequence-to-sequence frameworks. Further, input data may be bidirectionally encoded using both forward and backward separators. The forward and backward representations of the input data may be used to train the machine classifiers using a single generative model and/or shared parameters between the encoder and decoder of the machine classifier. During inference, the backward model may be used to reevaluate previously generated output sequences and the forward model may be used to generate an output sequence based on the previously generated output sequences.

Multi-turn dialogue response generation via mutual information maximization

A system includes one or more memory devices storing instructions, and one or more processors configured to execute the instructions to perform steps of a method. A method can include receiving observations and a corresponding class label, determining a candidate key based on the observations, determining a current memory state of a memory module based on a similarity of stored keys to the candidate key, generating a measurement vector based on the current memory state, concatenating the candidate key and measurement vector to form a state vector, determining, based on the state vector and an action distribution policy, an action of a plurality of actions such that the determined action maximizes an expected reduction in entropy as compared to the remaining actions of the plurality actions, executing the determined action, determining a value of the determined action, and updating, based on the value, the action distribution policy.

Systems providing a learning controller utilizing indexed memory and methods thereto

A system includes one or more memory devices storing instructions, and one or more processors configured to execute the instructions to perform steps of providing automated natural dialogue with a customer. The system may generate one or more events and commands temporarily stored in queues to be processed by one or more of a dialogue management device, an API server, and an NLP device. The dialogue management device may create adaptive responses to customer communications using a customer context, a rules-based platform, and a trained machine learning model.

Systems and methods for providing automated natural language dialogue with customers

Various embodiments may be generally directed to the use of an adversarial learning framework for persona-based dialogue modeling. In some embodiments, automated multi-turn dialogue response generation may be performed using a persona-based hierarchical recurrent encoder-decoder-based generative adversarial network (phredGAN). Such a phredGAN may feature a persona-based hierarchical recurrent encoder-decoder (PHRED) generator and a conditional discriminator. In some embodiments, the conditional discriminator may include an adversarial discriminator that is provided with attribute representations as inputs. In some other embodiments, the conditional discriminator may include an attribute discriminator, and attribute representations may be handled as targets of the attribute discriminator. The embodiments are not limited in this context.

Adversarial learning framework for persona-based dialogue modeling

The disclosed technology involves autonomously identifying goals and sub-goals from a user utterance and generating responses to the user based on the goals and sub-goals.

Systems and methods for generating automated natural language responses based on identified goals and sub-goals from an utterance

System and method of generating an executable action item in response to natural language dialogue are disclosed herein. A computing system receives a dialogue message from a remote client device of a customer associated with an organization, the dialogue message comprising an utterance indicative of an implied goal. A natural language processor of the computing system parses the dialogue message to identify one or more components contained in the utterance. The planning module of the computing system identifies the implied goal. The computing system generates a plan within a defined solution space. The computing system generates a verification message to the user to confirm the plan. The computing system transmits the verification message to the remote client device of the customer. The computing system updates an event queue with instructions to execute the action item according to the generated plan upon receiving a confirmation message from the remote client device.

Managing multifaceted, implicit goals through dialogue

Described are methods and systems are for generating dynamic conversational queries. For example, as opposed to being a simply reactive system, the methods and systems herein provide means for actively determining a user's intent and generating a dynamic query based on the determined user intent. Moreover, these methods and systems generate these queries in a conversational environment.

Systems and methods for generating dynamic conversational queries in user interfaces using state machine libraries

Methods and systems are described for generating dynamic interface options using machine learning models. The dynamic interface options may be generated in real time and reflect the likely goals and/or intents of a user. The machine learning model may provide these features by interpreting multi-modal feature inputs. For example, the machine learning model may include a first machine learning model, wherein the first machine learning model comprises a convolutional neural network, and a second machine learning model, wherein the second machine learning model performs a Weight of Evidence (WOE) analysis.

Systems and methods for generating dynamic conversational responses using ensemble prediction based on a plurality of machine learning models

In a variety of embodiments, machine classifiers may model multi-turn dialogue as a one-to-many prediction task. The machine classifier may be trained using adversarial bootstrapping between a generator and a discriminator with multi-turn capabilities. The machine classifiers may be trained in both auto-regressive and traditional teacher-forcing modes, with the generator including a hierarchical recurrent encoder-decoder network and the discriminator including a bi-directional recurrent neural network. The discriminator input may include a mixture of ground truth labels, the teacher-forcing outputs of the generator, and/or noise data. This mixture of input data may allow for richer feedback on the autoregressive outputs of the generator. The outputs can be ranked based on the discriminator feedback and a response selected from the ranked outputs.

Multi-turn dialogue response generation using asymmetric adversarial machine classifiers

Provided herein are systems and methods for using multi-modal regression to predict customer intent to contact a merchant. Multi-modal data including numerical data and unstructured data are extracted from customer interactions with the merchant. Features of the numerical data and the unstructured data are separately extracted and classified using techniques specific to the data types. The features for each type are then separately used to predict probabilities of customer intent. A neural network is used to combine the predictions into a single set of estimates of customer intent. This set of estimates of customer intents is used to estimate a probability that the customer will contact the merchant. The customer is then contacted based on the estimate.

Multi-modal regression to predict customer intent to contact a merchant

Systems and methods for generating responses to user input such as dialogues, and images are discussed. The system may generate, by a response generation module of at least one server, an optimal generated response to the user communication by applying an generative adversarial network. In some embodiments, the generative adversarial network may include a hierarchical recurrent encoder decoder generative adversarial network including a generator and a discriminator component.

Adversarial learning and generation of dialogue responses

Systems and methods are provided herein for autonomously determining and resolving a customer's perceived discrepancy during a customer service interaction. The method can include receiving an incoming communication from a customer; extracting a perceived state and an expected state (possibly of a product or service) based on the incoming communication; determining a perceived discrepancy between the perceived and expected states of the customer; retrieving customer information; extracting a current state of the customer from the retrieved customer information, verifying, by a rule-based platform, the discrepancy; generating a response based on the discrepancy after comparing the perceived stated with the current state, where the response may include a confirmation or a correction related to the discrepancy and a personalized explanation describing the current state of the customer; and outputting, for presentation to the customer, the response.

Systems and methods for automated discrepancy determination, explanation, and resolution with personalization

Memory augmented neural networks may use one or more neural encoders to transform input data into distributed representations and a memory module to store the representations with individual addresses. Memory augmented neural networks allow for few-shot learning capabilities because latent representations are persistent between training examples and gradient-based updates affect only certain memory locations via content-based lookups. When a query vector is not found in memory and the memory is full, existing memories that are positively associated with a particular representation may be identified, redundant memories may be aged, and updated memories may be generated. These updated memories retain relevant information acquired during training and reduce redundancy in the memories stored using the memory module, thereby improving the efficiency of data storage and reducing overfitting of data typically encountered with existing neural networks using memory modules.

Growing community of inventors

Chevy Chase, MD, United States of America

Erik Thomas Mueller