Embodiments herein provide a method for generating a real-time radio coverage map in a wireless network by a network apparatus. The method includes: receiving real-time geospatial information from one or more geographical sources in the wireless network; determining handover information of at least one user equipment (UE) in the wireless network from a plurality of base stations based on the real-time geospatial information; and generating the real-time radio coverage map based on the handover information of at least one UE and the real-time geospatial information.

A method for training an asymmetric generative adversarial network to generate an image and an electronic apparatus using the same are provided. The method includes the following. A first real image belonging to a first category, a second real image belonging to a second category and a third real image belonging to a third category are input to an asymmetric generative adversarial network for training the asymmetric generative adversarial network, and the asymmetric generative adversarial network includes a first generator, a second generator, a first discriminator and a second discriminator. A fourth real image belonging to the second category is input to the first generator in the trained asymmetric generative adversarial network to generate a defect image.

Methods include generating a scanner aerial image using a neural network, where the scanner aerial image is generated using a mask inspection image that has been generated by a mask inspection machine. Embodiments also include training the neural network with a set of images, such as with a simulated scanner aerial image and another image selected from a simulated mask inspection image, a simulated Critical Dimension Scanning Electron Microscope (CD-SEM) image, a simulated scanner emulator image and a simulated actinic mask inspection image.

The present disclosure relates to a method and an apparatus for deleting training data of a deep learning model. The trained data deleting method according to an exemplary embodiment of the present disclosure includes calculating a result value for a label allocated to data to be deleted which is included in the training data; reallocating a label of the data to be deleted by comparing the result value; generating a neutralized model obtained by neutralizing the deep learning model with the data to be deleted and a reallocated label of the data to be deleted as inputs; and training the neutral model based on retrained data which is training data, excluding the data to be deleted, among the trained data.

A method and an apparatus for retrieving a target are provided. The method may include: obtaining at least one image and a description text of a designated object; extracting image features of the image and text features of the description text by using a pre-trained cross-media feature extraction network; and matching the image features with the text features to determine an image that contains the designated object.

A device for machine learning is provided, including a first neural network layer, a second neural network layer with a normalization layer arranged in between. The normalization layer is configured to, when the device is undergoing training on a batch of training samples, receive multiple outputs of the first neural network layer for a plurality of training samples of the batch, each output comprising multiple data values for different indices on a first dimension and a second dimension; group the outputs into multiple groups based on the indices on the first and second dimensions; form a normalization output for each group which are provided as input to the second neural network layer. According to the application, the training of a deep convolutional neural network with good performance that performs stably at different batch sizes and is generalizable to multiple vision tasks is achieved, thereby improving the performance of the training.

Disclosed subject matter relates generally to forming a set of training parameters applicable to detection of two or more entities between and/or among a distribution of entities from a plurality of digitally stored observations. One or more training parameters of the set of training parameters may be modified to define a translation, which is applicable to detection of real-world entities corresponding to the two or more entities in the distribution of the digitally stored observations, wherein the forming of the translation is to be based, at least in part, on a first process to generate the two or more entities in the distribution of digitally stored observations and a second process to discriminate between and/or among the generated two or more entities based, at least in part, on the modified one or more training parameters

A training method of a generator network model and an electronic device for execution thereof are provided. The training method includes: extracting a first tensor matrix and a second tensor matrix, wherein the first tensor matrix and the second tensor matrix respectively represent a first picture and a second picture and individually include a plurality of first parameters and a plurality of second parameters; generating a plurality of third pictures according to a plurality of difference values between the first parameters of the first tensor matrix and the second parameters of the second tensor matrix; performing a similarity test on a plurality of original pictures and the plurality of third pictures; and adopting at least one of the third pictures whose similarity is lower than or equal to a similarity threshold as at least one new sample picture.

A method for training a machine learning system using conditionally independent training data includes receiving an input dataset (p(x, y, z)). A generative adversarial network, that includes a generator and a first discriminator, uses the input dataset to generate a training data (p.sub.s (x.sub.f, y.sub.f, z.sub.f)) by generating the values (x.sub.f, y.sub.f, z.sub.f). The first discriminator determines a first loss (L.sub.1) based on (x.sub.f, y.sub.f, z.sub.f) and (x, y, z). A divergence calculator modifies the training data based on a dependence measure (γ). The divergence calculator includes a second discriminator and a third discriminator. Modifying the training data includes receiving a reference value ({tilde over (y)}), and computing, by the second discriminator, a second loss (L.sub.2) based on (x.sub.f, y.sub.f, z.sub.f) and (x.sub.f, {tilde over (y)}, z.sub.f). The third discriminator computes a third loss (L.sub.3) based on (y.sub.f, z.sub.f) and ({tilde over (y)}, z.sub.f). Further, a fourth loss (L.sub.4) is computed based on L.sub.2 and L.sub.3. The training data is output from the generator if L.sub.1 and L.sub.4 satisfy a predetermined condition.

A method includes generating first electrocardiogram (ECG) data by adding synthetic noise to naturally occurring ECG data using a first deep neural network (DNN). The method further includes providing one of: (i) the first ECG data, or (ii) second ECG data including naturally occurring noise, to a second DNN. An output is generated by the second DNN indicating whether the second DNN received the first ECG data or the second ECG data.