International Business Machines Corporation

Syed Yousaf Shah

Petros Zerfos

Xuan-Hong Dang

Valentina Salapura

Ruchi Mahindru

Jing James Xu

Bei Chen

Dakuo Wang

Amos Cahan

Si Er Han

Ji Hui Yang

Dhavalkumar C Patel

Chuang Gan

Chatschik Bisdikian

Jun Song Wang

Peter Daniel Kirchner

Horst Cornelius Samulowitz

Long H Vu

Dominic Peter Harries

Gregory Bramble

Saket Sathe

Christopher R Gibson

Clifford A Pickover

Nancy Anne Greco

Jun Wang

Wesley M Gifford

Arunima Chaudhary

Abel Nicolas Valente

Daniel Karl I Weidele

Raji Lakshmi Akella

Voranouth Supadulya

John Dillon Eversman

Long Vu

Yunqi Guo

Milton Orlando Laverde Echeverria

Ashley Potter

Providing time-series forecasting by receiving target variable data and exogenous variable data, training a plurality of time-series models according to the target variable data and the exogenous variable data, determining a historical error for each of the plurality of time series models, and providing a time-series forecasting model having a lowest historical error.

Time series forecasting with exogenous variable data

To automate time series forecasting machine learning pipeline generation, a data allocation size of time series data may be determined based on one or more characteristics of a time series data set. The time series data may be allocated for use by candidate machine learning pipelines based on the data allocation size. Features for the time series data may be determined and cached by the candidate machine learning pipelines. Predictions of each of the candidate machine learning pipelines using at least the one or more features may be evaluated. A ranked list of machine learning pipelines may be automatically generated from the candidate machine learning pipelines for time series forecasting based upon evaluating predictions of each of the one or more candidate machine learning pipelines.

Automated time series forecasting pipeline generation

Embodiments relate to a system, program product, and method for employing deep learning techniques to fuse data across modalities. A multi-modal data set is received, including a first data set having a first modality and a second data set having a second modality, with the second modality being different from the first modality. The first and second data sets are processed, including encoding the first data set into one or more first vectors, and encoding the second data set into one or more second vectors. The processed multi-modal data set is analyzed, and the encoded features from the first and second modalities are iteratively and asynchronously fused. The fused modalities include combined vectors from the first and second data sets representing correlated temporal behavior. The fused vectors are then returned as output data.

Fusing multimodal data using recurrent neural networks

An assistive shopping cart system for guiding a user and monitoring the user's physical and cognitive conditions in a shopping environment is described. The assistive shopping cart system provides assistance and guidance to a user based on a user profile. The assistive shopping cart system routes the user through the shopping environment using a shopping route generated based on the user profile. The assistive shopping cart system also detects and tracks user physical and cognitive conditions and user actions in the shopping environment. The user actions are used to aid in the completion of a purchase/transaction and generate shopping notifications for the user and others to understand the process of the shopping experience.

Method, system, and medium for user guidance and condition detection in a shopping environment

Systems, computer-implemented methods, and computer program products to facilitate visualization of a model selection process are provided. According to an embodiment, a system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise an interaction backend handler component that obtains one or more assessment metrics of a model pipeline candidate. The computer executable components can further comprise a visualization render component that renders a progress visualization of the model pipeline candidate based on the one or more assessment metrics.

Visualization of a model selection process in an automated model selection system

Techniques regarding multivariate time series data analysis are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a time series analysis component that generates a machine learning model that discovers a dependency between multivariate time series data using an attention mechanism controlled by an uncertainty measure.

Determining multivariate time series data dependencies

Techniques regarding one or more automated machine learning processes that analyze time series data are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a time series analysis component that selects a machine learning pipeline for meta transfer learning on time series data by sequentially allocating subsets of training data from the time series data amongst a plurality of machine learning pipeline candidates.

Automated machine learning pipeline generation

Techniques are provided for the automated risk assessment of a document. In one embodiment, the techniques involve mapping, via a risk assessment engine, one or more sentences in a first document to one or more risk categories, identifying, via a classification engine, risk-associated language of the one or more sentences based on the one or more risk categories, mapping, via a risk assessment engine, the risk-associated language of the one or more sentences to one or more risk criterion of a risk criterion document, and generating, via a risk assessment engine, a first risk assessment based on the one or more risk criterion of the risk criterion document.

Automatic contract risk assessment based on sentence level risk criterion using machine learning

A system, method and program product for providing online privacy of image data. A centralized image privacy service is disclosed that includes: a user interface for allowing users to configure privacy profiles and provide profile images; an image scanning system that scans participating online sites for image data that matches the profile images; and a detection response system that determines a responsive action in response to a detected match based on an associated privacy profile, wherein the responsive action includes sending a masking request to the participating online site where the detected match occurred.

Privacy protection of images in online settings

A control unit is coupled to an infrared light source including at least an infrared light emitter. The infrared light source may be embedded in a wearable device. The control unit captures user preference and also receives current context associated with the wearable device and/or a user wearing the wearable device. The control unit determines automatically based on the current context meeting the user preference, whether to activate the infrared light source to emit infrared light or deactivate the infrared light source to stop emitting infrared light. Responsive to determining that the infrared light source is to be activated, the infrared light source is activated to emit the infrared light. Responsive to determining that the infrared light source is to be deactivated, the infrared light source is deactivated to stop emitting the infrared light.

Proactive image obfuscation

An emergency vehicle route management method, system, and non-transitory computer readable medium, include analyzing a map to plot roads, a user including user data, and an emergency vehicle including emergency vehicle data on the map, determining an intersection of the roads within a predetermined radius of the emergency vehicle, and delivering an action for the user based on a road on a path of the emergency vehicle being determined by the determining to intersect with a road of the user.

System, method and recording medium for emergency vehicle route guidance

Mechanisms are provided for generating a composite service. A request to generate the composite service is received that identifies a geospatial region of interest for the composite service. One or more types of components needed to generate the composite service are determined and, for each component of a plurality of components of the one or more types of components, a corresponding spatial coverage characteristic is determined. A subset of components, from the plurality of components, is selected based on the spatial coverage characteristics of the plurality of components and the geospatial region of interest. The composite service is then generated based on the selected subset of components from the plurality of components.

Providing a sensor composite service based on operational and spatial constraints

Growing community of inventors

Yorktown Heights, NY, United States of America

Syed Yousaf Shah