International Business Machines Corporation

Astellas Pharma Gmbh

Mari Abe Fukuda

Yasutaka Nishimura

Taku Sasaki

Asuka Unno

Shoichiro Watanabe

Kazuhito Akiyama

Hiroya Ogihara

Sanehiro Furuichi

Gaku Yamamoto

Taiga Nakamura

Kenichi Takasaki

Shun Jiang

Aly Megahed

Kugamoorthy Gajananan

Mark A Smith

Satoshi Hosokawa

Yuhko Kanoh Hasegawa

Junji Sato

Takuya Nakaike

Ken Ikeda

Masakatsu Kawakami

Ai Yoshino

Yoshihiro Keto

Takayuki Kushida

Jun-Jang Jeng

Teruo Koyanagi

Tsend Ochir Bat Ulzii

Yuka Sasaki

Reiya Takemura

Yohsuke Ozawu

A computer-implemented method, a computer program product, and a computer system for parallel cross validation in collaborative machine learning. A server groups local models into groups. In each group, each local device uses its local data to validate accuracies of the local models and sends a validation result to a group leader or the server. The group leader or the server selects groups whose variances of the accuracies are not below a predetermined variance threshold. In each selected group, the group leader or the server compares an accuracy of each local model with an average value of the accuracies and randomly selects one or more local models whose accuracies do not exceed a predetermined accuracy threshold. The server obtains weight parameters of selected local models and updates the global model based on the weight parameters.

Parallel cross validation in collaborative machine learning

A computational device maintains indications of a plurality of events associated with navigation of a plurality of vehicles in a geographical area. A determination is made as to whether to delete an event from the plurality of events, by performing: transmitting, by an event deletion manager, a query to a plurality of deletion determination agents on whether to delete the event; receiving, by the event deletion manager, an indication from the plurality of deletion determination agents whether to delete the event; and aggregating, by the event deletion manager, indications received from the plurality of deletion determination agents to determine whether to delete the event.

Deletion of events based on a plurality of factors in a connected car computing environment

A computer system trains a federated learning model. A federated learning model is distributed to a plurality of computing nodes, each having a set of local training data comprising labeled data samples. Statistical data is received from each computing node that indicates the node's count of data samples for each label, and is analyzed to identify one or more computing nodes having local training data in which a label category is underrepresented beyond a threshold value with respect to data samples. Additional data samples labeled with the underrepresented labels are provided, and the computing nodes perform training. Results of training are received and are processed to generate a trained global model. Embodiments of the present invention further include a method and program product for training a federated learning model in substantially the same manner described above.

Training federated learning models

A processor may generate, based on a predicted route of a user, a timeslot for presenting information to the user. The predicted route may be associated with a route segment, and the timeslot may be associated with the route segment. The processor may match one or more proposals to the timeslot associated with the route segment. The one or more proposals may be matched, at least in part, based on proposal criteria set by a proposal criteria user associated with the proposal. The processor may predict an acceptance potentiality of the user. The processor may select a subset of the one or more proposals. The processor may provide the selected subset of the one or more proposals. The selected subset may be provided, at least in part, based on the acceptance potentiality.

Route-based information slotting based on predicted acceptance

Utilizing a trained generative adversarial network (GAN) model to cause a computer to output multivariate forecasted time-series data by providing a trained GAN model, the GAN model comprising dilated convolutional layers for receiving time-series multivariate data, receiving time-series multivariable biometric data, generating, using the GAN model, successive time series multivariate biometric data according to the time-series multivariate biometric data, determining an outcome according to the successive time-series multivariate biometric data, and providing an output associated with the outcome.

Forecasting multivariate time series data

An embodiment of the invention may include a method, computer program product and computer system for managing mobile objects. The embodiment may manage, by a first computing system, a plurality of mobile objects moving within a geographic space. Managing the plurality of mobile objects may assisting with movement of the plurality of mobile objects. The embodiment may determine whether a first mobile object among the plurality of mobile objects is a real mobile object based on a first sensor information received from the first mobile object. The embodiment may use information received from the first mobile object in managing the plurality of mobile objects moving within the geographic space based on determining that the first mobile object is the real mobile object.

Management of mobile objects

A method, computer system, and a computer program product for preemptive collision mitigation is provided. The present invention may include calculating a future position of a first vehicle based on carprobe data from a first vehicle, wherein the carprobe data contains neural data of an operator of the first vehicle. The present invention may also include calculating a distance between the future position of the first vehicle and a future position of a second vehicle. The present invention may then include determining the calculated distance between the future position of the first vehicle and the future position of the second vehicle is below a threshold distance.

Preemptive mitigation of collision risk

One embodiment provides a method for augmenting missing values in historical or market data for deals. The method comprises receiving information relating to a set of deals. For any service included in one or more deals of the set of deals but not included in one or more other deals of the set of deals, the method further comprises augmenting, for any or all of the one or more other deals that does not include the service, one or more missing values for the service with one or more recommended values based on a recommendation algorithm. The service may be at any service level of a hierarchy of services.

Augmenting missing values in historical or market data for deals

A system and method for scheduling a business process including tasks, comprises a calculation unit, a determination unit, and a decision unit. The calculation unit is configured to calculate an estimated processing time required to execute the tasks. The determination unit is configured to calculate an estimated end time of a route including the tasks on the basis of the estimated processing time and schedule of a user to execute the tasks, and determine whether to apply speculative execution to the business process on the basis of the estimated end time. The decision unit is configured to decide to speculatively execute a task out of the tasks in the business process. The decision is made with reference to a remaining period for executing the task. The remaining period is calculated on the basis of a predicted execution timing of each task and a deadline of the business process.

Scheduling business process

One embodiment provides a method comprising receiving historic peer deals relating to at least one service, and a baseline and cost percentage estimation for each service. Historic peer cost data for each service is clustered to form at least one cluster. Each cluster includes similar unit costs, and has an assigned label. A classification model is trained based on each baseline received, each cost percentage estimation received, and each assigned label. For each assigned label, a corresponding probability distribution is computed based on the classification model. For each service of a new client solution, an assigned label for the service is predicted based on the classification model, and, based on a probability distribution corresponding to the assigned label predicted, transforming an initial range of historic peer cost data relating to the service into a narrower range for use in estimating a cost of the service with improved accuracy.

Effectiveness of service complexity configurations in top-down complex services design

Disclosed embodiments provide techniques for generation of vehicle travel data. Past-travel data including previously taken trips, along with metadata including, but not limited to, vehicle type, vehicle model, navigation device model, are obtained and classified. A travel route for simulation data is selected. Data from various previous trips are combined to create a simulation data set for the travel route. This data is based on actual vehicle data. This can enable a new level of accuracy for data analysis and simulations based on this data as compared with purely simulated data. Actual vehicle data accounts for driver behavior, road obstacles, signage, vegetation, vehicle characteristics, and other factors that may not be able to be captured with purely simulated traffic data. Using the data generated by disclosed embodiments enables these factors to be considered in traffic analysis.

System and method for generating vehicle travel data

An embodiment of the invention may include a method, computer program product and computer system for managing mobile objects. The embodiment may identify, by an event agent (EA), an event occurring in a geographic space in which a plurality of mobile objects move. The embodiment may include determining the event is an expected event based on predicting time-series changes of the event handled by the EA. The embodiment may manage the one mobile object based on the expected event.

Position data may gradually pseudonymized by a method, comprising: generating a sequence of relative positions from a sequence of absolute positions of a moving object; randomizing the sequence of relative positions using at least a sequence of random numbers generated from at least one seed; in response to receiving an analytical job comprising the at least one seed, restoring the sequence of relative positions from the randomized sequence of relative positions; and in response to receiving an analytical job comprising both the at least one seed and at least one absolute position derived from the sequence of absolute positions, restoring the sequence of absolute positions from the randomized sequence of relative positions.

Growing community of inventors

Tokyo, Japan

Mari Abe Fukuda