Abstract: A method comprising a deep learning based translation system translating a source sentence from language A to a target sentence in language B, placing constraints on the target sentence in language B.

Abstract: A computer-implemented method of generating a modified artificial neural network (ANN) from a base ANN having an ordered series of two or more successive layers of neurons, each layer passing data signals to the next layer in the ordered series, the neurons of each layer processing the data signals received from the preceding layer according to an activation function and weights for that layer comprises: detecting the data signals for a first position and a second position in the ordered series of layers of neurons; generating the modified ANN from the base ANN by providing an introduced layer of neurons to provide processing between the first position and the second position with respect to the ordered series of layers of neurons of the base ANN; deriving an initial approximation of at least a set of weights for the introduced layer using a least squares approximation from the data signals detected for the first position and a second position; and processing training data using the modified ANN to train the

Abstract: A computer-implemented method of generating a derived artificial neural network (ANN) from a base ANN comprises initialising a set of parameters of the derived ANN in dependence upon parameters of the base ANN; inferring a set of output data from a set of input data using the base ANN; quantising the set of output data; and training the derived ANN using training data comprising the set of input data and the quantised set of output data.

Abstract: A computer-implemented method of training an artificial neural network (ANN) by generating a first learned parameter for use in normalising input data values during a subsequent inference phase of the trained ANN. The method includes, for each of a series of batches of training data values, deriving a batch variance of the batch of training data values and a running variance of all training data values already processed in the training phase; generating an approximation of a current value of the first learned parameter so that a first scaling factor dependent upon the approximation of the first learned parameter and the running variance, is constrained to be equal to a power of two; and normalizing the batch of input data values by a second scaling factor dependent upon the approximation of the current value of the first learned parameter and the batch variance.

Abstract: An apparatus comprising circuitry that implements an artificial neural network training algorithm that uses weight tying.

Abstract: A method for personalizing a multimedia application includes receiving, from a module of the multimedia application, a request for a multimedia output, wherein the request is of a type; providing user data of a user of the multimedia application, the user data being representative of a personal taste of the user; and determining a multimedia output depending on the type and the user data.

Abstract: Embodiments of the invention relate to a method for video retrieval by providing a first audio video file, determining a first identifier of a first piece of music in the first audio video file, looking up for the first identifier first meta data in a music database, in which said first meta data are associated with said first identifier, providing second meta data of a second piece of music from the music database. The second piece of music is included in a second audio video file. Further the method is conducted by determining a similarity measure by comparing the first and second meta data, and providing the second audio video file or an identifier thereof depending on the similarity measure. Further embodiments relate to a server, a user device and a system and a computer program product for video retrieval.

Abstract: Embodiments of the invention relate to a method for video retrieval by providing a first audio video file, determining a first identifier of a first piece of music in the first audio video file, looking up for the first identifier first meta data in a music database, in which said first meta data are associated with said first identifier, providing second meta data of a second piece of music from the music database. The second piece of music is included in a second audio video file. Further the method is conducted by determining a similarity measure by comparing the first and second meta data, and providing the second audio video file or an identifier thereof depending on the similarity measure. Further embodiments relate to a server, a user device and a system and a computer program product for video retrieval.

Abstract: A method to determine a subgroup of users out of a group of users, comprising providing user profiles for each user of said group of users; providing real-time data of users of said group of users; determining said subgroup of users depending on said user profiles and said real-time data.

Abstract: A method for personalizing a multimedia application includes receiving, from a module of the multimedia application, a request for a multimedia output, wherein the request is of a type; providing user data of a user of the multimedia application, the user data being representative of a personal taste of the user; and determining a multimedia output depending on the type and the user data.