International Business Machines Corporation

University College Dublin

Jakub Marecek

Jonathan P Epperlein

Robert Shorten

Julien Monteil

Martin Mevissen

Giovanni Russo

Yassine Lassoued

Rodrigo Hernan Ordonez-Hurtado

Claudio Gambella

Wynita M Griggs

Michael Wurst

Pascal Pompey

Dimitrios Mavroeidis

Sergiy Zhuk

Rahul Nair

Randall Lee Cogill

Paulito Pedregosa Palmes

Fabian R Wirth

Bei Chen

Fearghal O'Donncha

Vincent Pa Lonij

Philipp Haehnel

Wynita May Griggs

Jonathan Peter Epperlein

Embodiments for identifying stochastic models representing the individual decision makers in a computing environment by a processor. One or more non-deterministic (stochastic, probabilistic) models may be identified according to a sequence of outcomes from decisions of each of a plurality of decision makers.

Identification of non-deterministic models of multiple decision makers

Embodiments for deep learning for partial differential equation (PDE)-based models by a processor. A trained forecasting model and consistency constraints may be generated using a PDE-based model, a discretization of the PDE-based model, historical inputs the of the PDE-based model, and a representation of consistency constraints to generate a predictive output.

Deep learning for partial differential equation (PDE) based models

In an approach for determining the illumination necessary along a vehicle's planned trajectory, a processor receives information relevant to the trajectory of a vehicle. A processor estimates the position of the vehicle at the particular time on the trajectory of the vehicle, using the data from the sensors. A processor estimates an illumination of the trajectory of the vehicle, using the model of the illumination of the road network and the data from the sensors, wherein a threshold is employed. A processor determines the threshold is exceeded. A processor changes a setting of the at least one headlight on the vehicle. A processor determines the threshold is exceeded. A processor alerts the user of the vehicle when adjustments to the at least one headlight that are necessary to illuminate the trajectory of the vehicle exceed physical limitations of the at least one headlight.

Adaptive headlights for the trajectory of a vehicle

A system and method for cognitive journey monitoring are presented. Embodiments comprise journey prediction, parsing of data sources, risk assessment and mitigation, and natural-language user interaction by a cognitive processor. Data is gathered from a plurality of data sources and analyzed in the context of one or more of the user's intention(s). A dialogue with the user, in natural language, aims to provide and select one or more suggestions relating to the one or more user intention(s) such that the risk(s) relating to the one or more user's intention(s) is reduced. During the dialogue, cognitive reasoning may be performed, wherein the cognitive reasoning includes the ability to justify each suggestion and the ability to infer information from the interaction such as, for example, data obtained in a dialogue may inform subsequent inferences. The embodiments may use speech synthesis and speech recognition in an interactive spoken dialogue.

Cognitive journey companion system

The cost of data-mining is estimated where data-mining services are delivered via a distributed computing system environment. System requirements are estimated for a particular data-mining task for an input data set having specified properties. Estimating system requirements includes applying a partial learning tool to operate on sample data from the input data set.

Estimating the cost of data-mining services

A method for forecasting time delays added to a scheduled start time and a scheduled end time of a task includes generating a stochastic model of the task and resources affecting the task, the stochastic model includes a reactionary delay component that is a function of previous task end times and a root cause delay component that is an independent random process at a specific time. The method further includes: calculating a probability distribution of time delays added to the scheduled start time as a combination of the reactionary delay component and the root cause delay component using the stochastic model to provide a probability distribution of start times; and calculating a probability distribution of time delays added to the scheduled end time as a combination of the reactionary delay component and the root cause delay component using the stochastic model to provide a probability distribution of end times.

Disruption forecasting in complex schedules

Embodiments for testing embedded systems and their applications in an Internet of Things (IoT) environment by a processor, denoted as a Hardware-in-the-Loop as a Service (HiLaaS). In a simulated environment, one or more simulated entities and one or more real entities in a networked system may be tested in real-time according to received control parameters, for a price. The price is estimated by the system, based on other parameters, and offered to the user to accept or reject. Alternatively, the user may specify the price, the system estimates control parameters, and the user can accept or reject the control parameters. One or more properties of the one or more entities, the network system, or combination thereof may be estimated based on the testing of the one or more simulated entities, when the price and control parameters are accepted.

Testing embedded systems and application using hardware-in-the-loop as a service (HILAAS)

A system and method for real-time pollution control at a traffic junction are presented. A pollution level may be determined at the traffic junction according to weather, traffic volume, traffic type, pollution measurements, topology, or a combination thereof. A traffic volume threshold may be determined for the traffic junction to maintain the pollution level below a pollution threshold. One or more parameters of the traffic junction may be set to maintain traffic volume below the traffic volume threshold.

Real-time pollution control at a traffic junction

A system and method for cognitive risk mitigation are presented. Embodiments comprise journey prediction, parsing of data sources, risk assessment and mitigation, and natural-language user interaction by a cognitive processor. Data is gathered from a plurality of data sources and analyzed in the context of one or more of the user's intention(s). A dialogue with the user, in natural language, aims to provide and select one or more suggestions relating to the one or more user intention(s) such that the risk(s) relating to the one or more user's intention(s) is reduced. During the dialogue, cognitive reasoning may be performed, wherein the cognitive reasoning includes the ability to justify each suggestion and the ability to infer information from the interaction such as, for example, data obtained in a dialogue may inform subsequent inferences. The embodiments may use speech synthesis and speech recognition in an interactive spoken dialogue.

Techniques for autonomously optimizing traffic flow amongst one or more traffic junctions are provided. In one example, a computer-implemented method can comprise generating, by a system operatively coupled to a processor, a piece-wise sinusoidal representation of traffic arrival at a first traffic junction. The computer-implemented method can also comprise determining, by the system, an offset a parameter of one or more traffic junctions based on the piece-wise sinusoidal representation and a polynomial objective.

Growing community of inventors

Dublin, Ireland

Jakub Marecek