Shutterstock, Inc.

Nokia Technologies Oy

Alcatel Lucent

Alessandra Sala

Harri Lindholm

Volker Hilt

Diego Lugones

Alice Marascu

Diarmaid Finnerty

Raúl Gómez Bruballa

Veronica Darling

Lauren Sarah Burnham-King

Emily Teoh

Raúl Gómez

Yue Jin

Jesus Omana Iglesias

Deepak Ajwani

Lei Jiao

Volker Frederich Hilt

Volker Friedrich Hilt

William Kennedy

Methods and systems for predicting performance of creative content are disclosed. Exemplary implementations may: receive a collection of images; provide a context to a user; serially cause display of pairs of images on a computer interface; receive user responses indicating which image of each pair is preferred given the context; determine a resonance value for each image based on a number of times the user responses indicate each image is preferred when displayed in a pair of images; determine a confidence score for each image; generate one or more models for predicting image performance based on one or more of the resonance value and the confidence score for each image; receive a plurality of candidate images; determine, using at least one model, a first metric set for each candidate; and cause display of a listing of the candidate images, the listing including the first metric set for each candidate image.

Predicting performance of creative content

A method for recommending results to a user from a search query is provided. The method includes receiving, in a search engine, a search query for a media file from a user, identifying a style preference of the user associated with a one or more media file attributes, based on a user-related search history, selecting, from a database, a one or more media files based on the search query and the style preference of the user, determining a style preference score for the one or more media files based on the media file attributes, and recommending to the user a top ranked media file based on the style preference score. A system including a memory storing instructions and one or more processors to execute the instructions to cause the system to perform the above method is also provided.

Media file recommendations for a search engine

A method for training a generative image model, including defining multiple sensitive categories and protected attributes associated with multiple training images, determining for a particular sensitive category a distribution of a protected attribute within the training images, and based on the distribution, calculating for each training image a corresponding image debiasing weight value associated with the protected attribute. The method further includes generating annotated training data including the training images, and for each training image, (1) the corresponding image debiasing weight value associated with the protected attribute and (2) a corresponding descriptive text caption. The method further includes performing a training process using the annotated training data to train a generative image model resulting in a trained model. A contribution of each image in the training images to an optimization loss of the training process is weighted during the training process using the corresponding image debiasing weight value.

Balanced generative image model training

An apparatus, method and computer program is described comprising: receiving interaction information via at least one user device, wherein the interaction information is related to at least one user using at least one user device in relation to a first content creation task; receiving sensor data relating to the at least one user from one or more sensors; and determining data, using a first model, the data comprising content creation performance data and user state data, wherein: the content creation performance data indicates performance of the at least one user in relation to the first content creation task, based, at least in part, on the interaction information and a first content created when the at least one user performs the first content creation task; and the user state data is based, at least in part, on the received sensor data in relation to the first content creation task.

User interaction

A configuration method, equipment, system and computer readable medium for determining a new configuration of calculation resources associated to a function and to at least one future time slot, the calculation resources being defined as different calculation capacities allocated to the function during a time slot and included within at least one computer, the configuration method comprising a first step of determining an old demand, a second step of determining a plurality of conditional probabilities associated to the old demand and to a plurality of new demands, a third step of determining a plurality of reconfiguration costs associated to a plurality available configurations, a step of selecting the new configuration as being the available configuration associated to the minimal reconfiguration cost.

Configuration method, equipment, system and computer readable medium for determining a new configuration of calculation resources

Growing community of inventors

Dublin, Ireland

Alessandra Sala