Wells Fargo Bank, N.a.

Bank of America Corporation

Jie Chen

Agus Sudjianto

Vijayan Narayana Nair

Weicheng Liu

William Anthony Nobili

Harsh Singhal

Sandi Setiawan

Timothy John Breault

Yiping Zhuang

Rao Fu

Shutian Zeng

Naveen Gururaja Yeri

Arun Pinto

Kevin D Oden

Aijun Zhang

Benhong Zhang

Zhe Zhang

Kaloyan Mihaylov

Meghan Alita Steach

Joel Vaughan

Jocelyn Sikora

Kiran Vuppu

Hungien Wang

Kurt Schieding

Daniel Hudson

Linwei Hu

Michael Chorba

Xiaoyu Liu

Carmel Nadav

Manish Pandey

Fernando Cela Diaz

Andrea Renee Turner

Bradford Timothy Winkelman

Peter B Vechnak

Michelle Warholic

Computing systems and technical methods that transform data structures and pierce opacity difficulties associated with complex machine learning algorithms are disclosed. Advances include a framework and techniques that include: i) global diagnostics; ii) locally interpretable models LIME-SUP-R and LIME-SUP-D; and iii) explainable neural networks.

Resolving opaqueness of complex machine learning applications

The innovation disclosed and claimed herein, in one aspect thereof, comprises systems and methods of validating models guided by machine learning algorithms. The innovation can begin by receiving a risk model for validation having multiple sets of data. A first data set is selected from as an input. Outputs are generated for validation. One output can be generating a second set of analysis results using a comparable algorithm to the risk model. Another output can be generating a second set of variables and transformations using a machine learning algorithm and an untransformed set of the selected variables to assess the set of selected transformations. Another output can be generating a third set of variables using one or more machine learning algorithms and an extended feature set of variables to assess the selected variables. The outputs are compared to the analysis results, coefficients, selected variables, and selected transformations. A report of the comparison is generated.

Model validation of credit risk

Systems, apparatuses, methods, and computer program products are disclosed for generating time series. A time series simulator receives information corresponding to a request for time series. The information is formatted into input data by the time series simulator. The input data comprises at least one continuous condition. A generator network of the continuous condition generative adversarial network (CCGAN) generates the time series based directly on a value of the at least one continuous condition. The time series is provided such that the time series is at least one of (a) provided as input to an analysis pipeline or (b) received by a user computing device wherein a representation of at least a portion of the one or more time series is provided via an interactive user interface of the user computing device.

Systems and methods for time series simulation

Dynamic predictive modelling is disclosed. A subset of joined pairs can be selected from a table of joined pairs in which each pair includes snapshot time and forecast time, the joined pairs form a forecasting triangle that spans predetermined increments of time, and the subset of joined pairs comprises a vector in the forecasting triangle. A predictive model can be utilized to operate over the data associated with the subset of joined pairs and produce a prediction. The subset of joined pairs can be dependent on the predictive model. In one instance, data associated with the subset of pairs can be loaded into processor cache to facilitate efficient processing.

Dynamic prediction modeling

The concept involves efficiently using machine learning to quickly identify possible fraudulent applications in small business loan and credit applications by automatically flagging applications that meet certain criteria. In one preferred implementation, the tool compares a business description to a selected NAICS code in a loan application to assess the potential for fraud. Specifically, an algorithm can match the leftmost two digits of the selected code with the description of the category from an applicant. An engine calculates a probability of a fraud score based on the matching attached to the application. Because the tool detects fraud proactively rather than reactively, it substantially reduces computational costs and resources and reduces the biases associated with highly intensive manual work.

Detecting fraud in credit applications

Computing systems and technical methods that transform data structures and pierce opacity difficulties associated with complex machine learning modules are disclosed. Advances include a framework and techniques that include: i) global diagnostics; ii) locally interpretable models LIME-SUP-R and LIME-SUP-D; and iii) explainable neural networks. Advances also include integrating LIME-SUP-R and LIME-SUP-D approaches that create a transformed data structure and replicated modeling over local and global effects and that yield high interpretability along with high accuracy of the replicated complex machine learning modules that make up a machine learning application.

The innovation disclosed and claimed herein, in one aspect thereof, comprises systems and methods of validating models guided by machine learning algorithms. The innovation can begin by receiving a risk model for validation having multiple sets of data. A first data set is selected from as an input. Outputs are generated for validation. One output can be generating a second set of analysis results using a comparable algorithm to the risk model. Another output can be generating a second set of variables and transformations using a machine learning algorithm and an un-transformed set of the selected variables to assess the set of selected transformations. Another output can be generating a third set of variables using one or more machine learning algorithms and an extended feature set of variables to assess the selected variables. The outputs are compared to the analysis results, coefficients, selected variables, and selected transformations. A report of the comparison is generated.

Systems, methods, and other embodiments associated with dynamic predictive modelling. According to one embodiment, a method includes creating a joined pair including a snapshot time and the forecast time. The joined pair is stored in the storage device. A subset of data associated with the joined pair is selected from data stored in the storage device. The subset of data is selected based, at least in part, on predetermined increments of time between the snapshot time and the forecast time. The snapshot time, the forecast time, and the subset of data is provided to a processing device from the storage device.

A data driven and forward looking risk and reward appetite methodology for consumer and small business is described. The methodology includes account level historical data collection for customers associated with accounts as part of a portfolio. The account level historical data is segmented into groups of customers with similar revenues and loss characteristics. Segmented data is decomposed into seasoning, vintage, and cycle effects. Statistical clusters are formed based upon the data and effects. A simulation is applied to the statistical clusters and prediction data is generated. A simulation strategy to forecast and simulate revenue and loss volatility is developed. Efficient frontier curves of risk (e.g., return volatility) and reward (e.g., expected return) are created for the current portfolio under various economic scenarios.

Risk and reward assessment mechanism

Embodiments of the present invention relate to systems, methods and computer program products that model consumer leverage and provide a leading indicator that predicts increases or decreases in consumer net non-collectibles. To do so, for example, the present invention determines the growth of consumers' spending and borrowing, and tracks a relationship between the value of a ratio that compares consumers' spending and borrowing and the value of the equilibrium of the ratio that compares consumers' spending and borrowing. This relationship is then applied to predict changes in consumers' ability to repay borrowed funds and consumer net non-collectibles.

Consumer leverage modeling

Embodiments of the present invention relate to methods and apparatuses for determining leading indicators and/or for modeling one or more time series. For example, in some embodiments, a method is provided that includes: (a) receiving first data indicating the value of a total income amount for a plurality of consumers over a period of time; (b) receiving second data indicating the value of a total debt amount for a plurality of consumers over a period of time; (c) selecting a consumer leverage time series that compares the total income amount to the total debt amount over a period of time; (d) modeling the consumer leverage time series based at least partially on the first and second data; (e) determining, using a processor, the value of the cycle component for a particular time; and (f) outputting an indication of the value of the cycle component for the particular time.

Determining leading indicators

Embodiments of the present invention evaluate consumer spending and borrowing patterns and, based thereon, forecast changes in consumer failure to repay rates. Embodiments of the present invention then develop macroeconomic variables that reflect the forecasted changes in consumer failure to repay rates and implement those macroeconomic variables into credit-approval decision models. The implemented macroeconomic variables adjust the decision models' credit-approval thresholds to account for the forecasted changes in consumer failure to repay rates. For example, if forecasts indicate decreasing credit failure to repay rates, then macroeconomic variables are developed and implemented in decision models to reduce credit-approval thresholds, thereby reducing qualifying creditworthiness scores and making it easier to get credit. On the other hand, for example, if forecasts indicate increasing credit failure to repay rates, then macroeconomic variables are developed and implemented in decision models to increase credit-approval thresholds, thereby restricting access to credit and reducing future losses from consumer failures to repay.

Credit-approval decision models

A data driven and forward looking risk and reward appetite methodology for consumer and small business is described. The methodology includes customer segmentation to create pools of homogeneous assets in terms of revenue and loss characteristics, forward looking simulation to forecast expected values and volatilities of revenue and loss, and risk and reward optimization of the portfolio. One methodology used for modeling revenue and loss is a generalized additive effect decomposition model to fit historical data. Based on the model, a segmentation procedure is performed, which allows for creation of groups of customers with similar revenue and loss characteristics. An estimation procedure for the model is developed and a simulation strategy to forecast and simulate revenue and loss volatility is developed. Efficient frontier curves of risk (e.g., return volatility) and reward (e.g., expected return) are created for the current portfolio under various economic scenarios.

Growing community of inventors

Charlotte, NC, United States of America

Jie Chen