Abstract: Software and lasers are used in finishing apparel to produce a desired wear pattern or other design. A technique includes using machine learning to create or extract a laser input file for wear pattern from an existing garment. Machine learning can be by a generative adversarial network, having generative and discriminative neural nets. The generative adversarial network is trained and then used to create a model. This model is used generate the laser input file from an image of the existing garment with the finishing pattern. With this laser input file, a laser can re-create the wear pattern from the existing garment onto a new garment.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for processing an input data element to generate a prediction output that characterizes the input data element. In one aspect, a method comprises: determining a respective attention weight between an input data element and each of a plurality of reference data elements; processing each of the reference data elements using the encoder neural network to generate a respective value embedding of each reference data element; determining a combined value embedding of the reference data elements based on (i) the respective value embedding of each reference data element, and (ii) the respective attention weight between the input data element and each reference data element; and processing the combined value embedding of the reference data elements using a prediction neural network to generate the prediction output that characterizes the input data element.

Abstract: A model configuration selection system, the model configuration selection system comprising a processing circuitry configured to: (A) obtain: (a) one or more model configurations, each model configuration includes a set of parameters utilized to generate respective models, and (b) a training data-set comprising a plurality of unlabeled records, each unlabeled record including a collection of features describing a given state of a physical entity; (B) cluster the training data-set into two or more training data-set clusters using a clustering algorithm; (C) label (a) the unlabeled records of a subset of the training data-set clusters with a synthetic normal label, giving rise to a normal training data-set, and (b) the unlabeled records of the training data-set clusters not included in the subset with a synthetic abnormal label; (D) train, for each model configuration, using the normal training data-set, a corresponding model utilizing the corresponding set of parameters, each model capable of receiving the unl

Abstract: Quantized softmax layers in neural networks are described. Some embodiments involve receiving, at an input to a softmax layer of a neural network from an intermediate layer of the neural network, a non-normalized output comprising a plurality of intermediate network decision values. Then for each intermediate network decision value of the plurality of intermediate network decision values, the embodiment involves: calculating a difference between the intermediate network decision value and a maximum network decision value; requesting, from a lookup table, a corresponding lookup table value using the difference between the intermediate network decision value and the maximum network decision value; and selecting the corresponding lookup table value as a corresponding decision value. A normalized output is then generated comprising the corresponding lookup table value for said each intermediate network decision value of the plurality of intermediate network decision values.

Abstract: Methods and apparatuses are described for automatic refinement of intent classification for virtual assistant applications. A computing device creates utterance groups comprising messages. The device determines, using a first machine learning (ML) classifier, a first predicted intent associated with each utterance group and determines, using a second ML classifier, a second predicted intent associated with each utterance group. The device combines the first and second predicted intents to generate a final predicted intent, determines, for each incomprehensible utterance group, whether the final predicted intent overlaps with the final predicted intent for other utterance groups, and selects messages having no overlapping intent to create new intents.

Abstract: A mechanism is provided in a data processing system for preventing data leakage in automated machine learning. The mechanism receives a data set comprising a label for a target variable for a classifier machine learning model and a set of features. For each given feature in the set of features, the mechanism trains a subprime classifier model using the given feature as a target variable and remaining features as independent input features, tests the subprime classifier model, and records results of the subprime classifier model. The mechanism performs statistical analysis on the recorded results to identify an outlier result corresponding to an outlier subprime classifier model.

Abstract: The present disclosure relates to system(s) and method(s) for tuning an analytical model. The system builds a global analytical model based on modelling data received from a user. Further, the system analyses a target eco-system to identify a set of target eco-system parameters. The system further selects a sub-set of model parameters, corresponding to the set of target eco-system parameters, from a set of model parameters. Further, the system generates a local analytical model based on updating the global analytical model, based on the sub-set of model parameters and one or more PMML wrappers. The system further deploys the local analytical model at each node, from a set of nodes, associated with the target eco-system. Further, the system gathers test results from each node based on executing the local analytical model. The system further tunes the sub-set of model parameters associated with the local analytical model using federated learning algorithms.

Abstract: An artificial neural network (ANN) quantization method for generating an output ANN by quantizing an input ANN includes: obtaining second parameters by quantizing first parameters of the input ANN; obtaining a sample distribution from an intermediate ANN in which the obtained second parameters have been applied to the input ANN; and obtaining a fractional length for the sample distribution by quantizing the obtained sample distribution.

Abstract: An apparatus for optimizing a computational network is configure to receive an input at a first processing component. The first processing component may include at least a first programmable processing component and a second programmable processing component. The first programmable processing component is configured to compute a first nonlinear function and the second programmable processing component is configured to compute a second nonlinear function which is different than the second nonlinear function. The computational network which may be a recurrent neural network such as a long short-term memory may be operated to generate an inference based at least in part on outputs of the first programmable processing component and the second programmable processing component.

Abstract: A computing system is configured to access intermediate outputs of a neural network by augmenting a data flow graph generated for the neural network. The data flow graph includes a plurality of nodes interconnected by connections, each node representing an operation to be executed by the neural network. To access the intermediate output, the data flow graph is augmented by inserting a node representing an operation that saves the output of a node which produces the intermediate output. The node representing the save operation is inserted while maintaining all existing nodes and connections in the data flow graph, thereby preserving the behavior of the data flow graph. The augmenting can be performed using a compiler that generates the data flow graph from program code.

Abstract: Functions are added to a deep neural network (“DNN”) computation graph for encoding data structures during a forward training pass of the DNN and decoding previously-encoded data structures during a backward training pass of the DNN. The functions added to the DNN computation graph can be selected based upon on the specific layer pairs specified in the DNN computation graph. Once a modified DNN computation graph has been generated, the DNN can be trained using the modified DNN computation graph. The functions added to the modified DNN computation graph can reduce the utilization of memory during training of the DNN.

Abstract: From a quantum program a first mutant is generated using a processor and a memory, where the first mutant is a randomly-generated transformation of the quantum program. A quality score, a correctness distance, and a probability of acceptance corresponding to the first mutant are computed. An acceptance corresponding to the first mutant is determined according to the probability of acceptance. Upon determining that an acceptance of the first mutant corresponding to the probability of acceptance exceeds an acceptance threshold, the quantum program is replaced with the first mutant. Upon determining that the quality score exceeds a storage threshold and that the correctness distance is zero, the first mutant is stored. These actions are iterated until reaching an iteration limit.

Abstract: Systems and computerized methods for determining patterns in user activity such that user contextual information can be provided based on the patterns. Historic data associated with a first user is received and entity information is determined based on the historic data. Current location data associated with the first user is received and prediction information is generated based on a combination of the entity information and the current location data for a current time period. Contextual insight information is determined based on a combination of the prediction information with at least one of the current location data, real time data, and contextual data, the contextual insight information including a recommendation of at least one of an activity, a purchase, and an accessory associated with the current location of the first user.

Abstract: An electronic apparatus is provided. The electronic apparatus includes: a memory storing a trained model including a plurality of layers; and a processor initializing a parameter matrix and a plurality of split variables of a trained model, calculating a new parameter matrix having a block-diagonal matrix for the plurality of split variables and the trained model to minimize a loss function for the trained model, a weight decay regularization term, and an objective function including a split regularization term defined by the parameter matrix and the plurality of split variables, vertically splitting the plurality of layers according to the group based on the computed split parameters and reconstruct the trained model using the computed new parameter matrix as parameters of the vertically split layers.

Abstract: Methods and systems for performing diagnostic functions for a deep learning model are provided. One system includes one or more components executed by one or more computer subsystems. The one or more components include a deep learning model configured for determining information from an image generated for a specimen by an imaging tool. The one or more components also include a diagnostic component configured for determining one or more causal portions of the image that resulted in the information being determined and for performing one or more functions based on the determined one or more causal portions of the image.

Abstract: A contact center analysis system can receive various types of communications from customers, such as audio from telephone calls, voicemails, or video conferences; text from speech-to-text translations, emails, live chat transcripts, text messages, and the like; and other media or multimedia. The system can segment the communication data using temporal, lexical, semantic, syntactic, prosodic, user, and/or other features of the segments. The system can cluster the segments according to one or more similarity measures of the segments. The system can use the clusters to train a machine learning classifier to identify one or more of the clusters as waypoints (e.g., portions of the communications of particular relevance to a user training the classifier). The system can automatically classify new communications using the classifier and facilitate various analyses of the communications using the waypoints.

Abstract: Methods and systems are provided to determine suitable hyperparameters for a machine learning model and/or feature engineering process. A suitable machine learning model and associated hyperparameters are determined by analyzing a dataset. Suitable hyperparameter values for compatible machine learning models having one or more hyperparameters in common and a compatible dataset schema are identified. Hyperparameters may be ranked according to each of their respective influences on a model performance metrics, and hyperparameter values identified as having greater influence may be more aggressively searched.

Abstract: Methods, systems, and apparatuses for implementing advanced content retrieval are described. Machine learning methods may be implemented so that a system may predict when a user device may experience network disconnections. The system may also predict the type of content one or more applications on the user device may seek to download during the network disconnection period. Neural networks may be trained based on user activity log data and may implement machine-learning techniques to determine user preferences and settings for advanced content retrieval. The system may predict when a user may want to download content in advance, the type of content the user may be interested in, anticipated network connectivity, and anticipated battery consumption. The system may then generate recommendations for the user device based on the predictions. If a user agrees with the recommendations, the system may obtain and cache the content.

Abstract: A system and method of configuring a graphical control structure for controlling a machine learning-based automated dialogue system includes configuring a root dialogue classification node that performs a dialogue intent classification task for utterance data input; configuring a plurality of distinct dialogue state classification nodes that are arranged downstream of the root dialogue classification node; configuring a graphical edge connection between the root dialogue classification node and the plurality of distinct state dialogue classification nodes that graphically connects each of the plurality of distinct state dialogue classification nodes to the root dialogue classification node, wherein (i) the root dialogue classification node, (ii) the plurality of distinct classification nodes, (iii) and the transition edge connections define a graphical dialogue system control structure that governs an active dialogue between a user and the machine learning-based automated dialogue system.

Abstract: Disclosed is an acquisition method for domain rule knowledge of an industrial process. The method comprises the steps of: establishing a domain rule base, establishing a semantic knowledge base, and combining the domain rule base and the semantic knowledge base so as to realize an augmented update of a domain rule knowledge base; describing the domain knowledge of the industrial process by using weighted first-order logic rules so as to form a training sample set of the first-order logic rules; performing a weight learning by applying probability soft logic and the training sample set of the first-order logic rules so as to realize weight to non-weighted rules; performing rule learning through a machine learning algorithm so as to obtain a first-order logic rule on a change in optimization decision-making semantic when multi-source data semantic information changes.