International Business Machines Corporation

Amazon Technologies, Inc.

Google Inc.

University of Minnesota

Ramesh Natarajan

Soumyadip Ghosh

Jonathan R Hosking

Xiaoxuan Zhang

Shivaram Subramanian

Edwin Peter Dawson Pednault

Jonathan R M Hosking

Shivaram Subramaniam

Ning Duan

David Nahamoo

Lior Horesh

Chai Wah Wu

Raya Horesh

Christopher Wildsmith

Peder Andreas Olsen

Se June Hong

Thomas Phan

Daniel Patrick Connors

Sholom M Weiss

Amol N Ghoting

Arindam Banerjee

Gregory Michael Duncan

Michael Widman

Satoki Mitsumori

Hanhuai Shan

Lin Li

Yiwei Chen

Michael Kochte

Jonathan Hosking

Jonathan Richard Morley Hosking

Kamalakannan Elangovan

James Kingsbery

Sravan Kumar Kasturi

Methods and systems for determining control inputs to a manufacturing apparatus to manufacture a product are described. A processor may receive model data including initial state data indicating an initial state of an input material, a set of model control inputs, and target measurement data associated with a target product. The processor may learn a causal predictive model based on the target data. Each state of the causal predictive model may be based on an application of the model control inputs on a previous state of the causal predictive model. The processor may compare a final state of the causal predictive model with the target measurement data to determine a difference. The processor may determine, based on the difference, a set of control inputs to be assigned to one or more controls. The one or more controls may define a design of the manufacturing process of an end product.

Free-form production based on causal predictive models

The present disclosure provides various transformations to be used in analysis of a large number of transactions to detect anomalies that would indicate potential fraudulent or criminal activity. Such transformations may be applied, for example, using a machine learning system. According to some examples, each of various transformations may be used to detect a particular type of behavioral anomaly. When multiple disparate transformations are considered together by the machine learning system, anomalous activity related to potential fraudulent or criminal activity can be detected more frequently and with greater accuracy.

Method for identifying anomalous transactions using machine learning

Devices and techniques are generally described for determining a recalibration frequency of a state space model. In various examples, a first hyperparameter for a first dataset may be determined. A residual value between a first data point of the first dataset and a machine learning model fitted to the first dataset may be determined. A plurality of second datasets may be generated based on the residual value. Second hyperparameters may be determined for the plurality of second datasets. A variability of the second hyperparameters may be determined. A third hyperparameter may be determined for a subset of the first dataset. A recalibration frequency may be determined for the machine learning model by comparing the third hyperparameter to the variability of the second hyperparameters.

Recalibration frequency determination for state space models

A data set comprising records of state change events of items of an item collection, as well as records of asynchronous operations associated with the items, is obtained. The numbers of records in the data set may differ from one item to another. Using the data set, a Bayesian forecasting model employing a deconvolution algorithm is trained. The model generates estimates of metrics of a type of asynchronous operation using a combination of a category-level distribution of the asynchronous operation, an item-level distribution, and a category-versus item adjustment. A trained version of the model is stored.

Combining group-level and item-level information for forecasting by empirical Bayesian deconvolution based technique

Software that performs the following steps: (i) receiving data from a first database and data from a second database, (ii) identifying a training subset and a test subset from the received data; (iii) generating a first graphical model using data from the training subset; (iv) generating a second graphical model using data from the training subset; (v) determining respective weights for the first graphical model and the second graphical model by using an expectation maximization method on data from the test subset; (vi) generating a third graphical model by interpolating at least the first graphical model and the second graphical model using their respectively determined weights; and (vii) defining one or more links between the data from the first database and the data from the second database using the third graphical model.

Generating an expected prescriptions model using graphical models

A predictive-control approach allows an electricity provider to monitor and proactively manage peak and off-peak residential intra-day electricity usage in an emerging smart energy grid using time-dependent dynamic pricing incentives. The daily load is modeled as time-shifted, but cost-differentiated and substitutable, copies of the continuously-consumed electricity resource, and a consumer-choice prediction model is constructed to forecast the corresponding intra-day shares of total daily load according to this model. This is embedded within an optimization framework for managing the daily electricity usage. A series of transformations are employed, including the reformulation-linearization technique (RLT) to obtain a Mixed-Integer Programming (MIP) model representation of the resulting nonlinear optimization problem. In addition, various regulatory and pricing constraints are incorporated in conjunction with the specified profit and capacity utilization objectives.

Managing time-substitutable electricity usage using dynamic controls

Systems and methods for modeling and forecasting cyclical demand systems in the presence of dynamic control or dynamic incentives. A method for modeling a cyclical demand system comprises obtaining historical data on one or more demand measurements over a plurality of demand cycles, obtaining historical data on incentive signals over the plurality of demand cycles, constructing a model using the obtained historical data on the one or more demand measurements and the incentive signals, wherein constructing the model comprises specifying a state-space model, specifying variance parameters in the model, and estimating unknown variance parameters.

System and method for modeling and forecasting cyclical demand systems with dynamic controls and dynamic incentives

Systems are provided to estimate autoregressive moving average (ARMA) models using maximum likelihood estimation and analytical derivatives, and to use such models for forecasting. The evaluation of the analytical derivatives during estimation of the model parameters may be performed using a state space representation with certain characteristics. An ARMA model estimated using maximum likelihood estimation, analytical derivatives, and the state space representation with certain characteristics can be used to forecast/predict values that are likely to occur in the future, given some set of previously-occurring values.

Analytical derivative-based ARMA model estimation

Corresponding to each forecasting model of a family of related models for a time series sequence, a respective state space representation is generated. One or more cross-validation iterations are then executed for each model of the family. In a given iteration, a training variant of the time series sequence is generated, with a subset of the time series sequence entries replaced by representations of missing values. Predictions for the missing values are obtained using the state space representation and the training variant, and a model quality metric is obtained based on prediction errors. The optimal model of the family is selected using the model quality metrics obtained from the cross validation iterations.

Selecting forecasting models for time series using state space representations

A streams platform is used. Multiple streams of electricity usage data are received, each from an electrical meter providing periodic updates to electrical usage for devices connected to the electrical meter. Weather information is received corresponding to locations where the electrical meters are. Real-time predictive modeling of electricity demand is performed based on the received multiple streams of electricity usage data and the received weather information, at least by performing: updating a state space model for electrical load curves using the usage data from the streams and the weather, wherein the updating uses current load observations for the multiple streams for a current time period; and creating forecast(s) for the electricity demand. The forecast(s) of the electricity demand are output. Appliance-level predictions may be made and used, and substitution effects and load management functions may be performed.

Real-time forecasting of electricity demand in a streams-based architecture with applications

The computer creates a utility demand forecast model for weather parameters by receiving a plurality of utility parameter values, wherein each received utility parameter value corresponds to a weather parameter value. Determining that a range of weather parameter values lacks a sufficient amount of corresponding received utility parameter values. Determining one or more utility parameter values that corresponds to the range of weather parameter values. Creating a model which correlates the received and the determined utility parameter values with the corresponding weather parameters values.

Utility usage forecasting

The invention pertains to a system and method for a set of middleware components for supporting the execution of computational applications on high-performance computing platform. A specific embodiment of this invention was used to deploy a financial risk application on Blue Gene/L parallel supercomputer. The invention is relevant to any application where the input and output data are stored in external sources, such as SQL databases, where the automatic pre-staging and post-staging of the data between the external data sources and the computational platform is desirable. This middleware provides a number of core features to support these applications including for example, an automated data extraction and staging gateway, a standardized high-level job specification schema, a well-defined web services (SOAP) API for interoperability with other applications, and a secure HTML/JSP web-based interface suitable for non-expert and non-privileged users.

System and method for automating and scheduling remote data transfer and computation for high performance computing

A system, method and computer program product provides for multiple imputation of missing data elements in retail data sets used for modeling and decision-support applications based on the multi-dimensional, tensor structure of the data sets, and a fast, scalable scheme is implemented that is suitable for large data sets. The method generates multiple imputations comprising a set of complete data sets each containing one of a plurality of imputed realizations for the missing data values in the original data set, so that the variability in the magnitudes of these missing data values can be captured for subsequent statistical analysis. The method is based on the multi-dimensional structure of the retail data sets incorporating tensor factorization, that in a preferred embodiment can be implemented using fast, scalable imputation methods suitable for large data sets, to obtain multiple complete data sets in which the original missing values are replaced by various imputed values.

Growing community of inventors

Pleasantville, NY, United States of America

Ramesh Natarajan