Abstract: Systems, computer program products, and methods are described herein for de-conflicting data labeling in real-time deep learning systems. The present invention is configured to retrieve one or more dynamically generated expert profiles; and determine an optimal expert mix of experts to classify the transaction into a transaction types, wherein the expert profiles comprises: (i) shared information metrics, (ii) divergence metrics, (iii) characteristics associated with the one or more experts, (iv) a predictive accuracy of the one or more experts, (v) an exposure score associated with the one or more experts, and (vi) information associated with the transaction, wherein the optimal expert mix comprises: (i) a best expert for classifying the transaction, (ii) a combination score from at least the portion of the one or more experts evaluating the transaction simultaneously, and (iii) a sequence of at least the portion of the one or more experts analyzing the transaction.

Abstract: A condition determination system includes: an operation condition data obtaining unit that obtains operation condition data indicating an operation condition of a facility; and a determination unit that determines, based on the operation condition data, a level of a phenomenon that occurs due to the operation condition of the facility.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for few-shot learning-based generator training are disclosed. An exemplary method may start with obtaining a teacher model and a plurality of training samples, as well as a generator for generating more training samples. After generating a plurality of additional training samples using the method may continue with feeding the plurality of generated additional training samples into the teacher model to obtain a plurality of first statistics; and feeding the plurality of training samples into the teacher model to obtain a plurality second statistics. Then the method further includes training the generator to minimize a distance between the plurality of first statistics and the plurality of second statistics.

Abstract: An explainable transducer transformer (XTT) may be a finite state transducer, together with an Explainable Transformer. Variants of the XTT may include an explainable Transformer-Encoder and an explainable Transformer-Decoder. An exemplary Explainable Transducer may be used as a partial replacement in trained Explainable Neural Network (XNN) architectures or logically equivalent architectures. An Explainable Transformer may replace black-box model components of a Transformer with white-box model equivalents, in both the sub-layers of the encoder and decoder layers of the Transformer. XTTs may utilize an Explanation and Interpretation Generation System (EIGS), to generate explanations and filter such explanations to produce an interpretation of the answer, explanation, and its justification.

Abstract: An apparatus to facilitate neural network (NN) training is disclosed. The apparatus includes training logic to receive one or more network constraints and train the NN by automatically determining a best network layout and parameters based on the network constraints.

Abstract: An artificial neuron including: a membrane capacitor; an input of an external synaptic excitation in current, the membrane capacitor integrating the input current; a negative-feedback impulse circuit, supplied by a power supply at a negative voltage between ?200 mV and 0 mV and at a positive voltage between 0 mV and +200 mV, including: a bridge based on pMOS and nMOS transistors in series and linked by a midpoint to the membrane capacitor, the midpoint defining the output of the artificial neuron, at least one delay capacitor between the gate and the source of one of the transistors of the bridge, at least two CMOS inverters between the membrane capacitor and the gates of the transistors of the bridge.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for few-shot learning-based generator training based on raw data collected from a specific domain or class. In cases where the raw data is collected from multiple domains but is not easily divisible into classes, the invention describes training multiple generators based on a pivot-sample-based training process. Pivot samples are randomly selected from the raw data for clustering, and each cluster of raw data may be used to train a generator using the few-shot learning-based training process.

Abstract: A sequential data analysis apparatus extracts a pattern of two or more sets of items based on an appearance frequency of each of different sets of items in first sequential data, selects a pattern of two or more sets of items based on an appearance frequency of a sub-pattern formed of a portion of the extracted pattern, creates a related pattern including the same last set of items as and the other sets of items different from the selected characteristic pattern, calculates an evaluation value of the related pattern, creates a prediction model by organizing data of the characteristic pattern and the related pattern, and applies second sequential data to the prediction model to determine a result which the second sequential data is likely to lead to.

Abstract: A disclosed neural network processing system includes a host computer system, a RAMs coupled to the host computer system, and neural network accelerators coupled to the RAMs, respectively. The host computer system is configured with software that when executed causes the host computer system to write input data and work requests to the RAMS. Each work request specifies a subset of neural network operations to perform and memory locations in a RAM of the input data and parameters. A graph of dependencies among neural network operations is built and additional dependencies added. The operations are partitioned into coarse grain tasks and fine grain subtasks for optimal scheduling for parallel execution. The subtasks are scheduled to accelerator kernels of matching capabilities. Each neural network accelerator is configured to read a work request from the respective RAM and perform the subset of neural network operations on the input data using the parameters.

Abstract: A method for using machine learning techniques to analyze past decisions made by administrators concerning account opening requests and to recommend whether an account opening request should be allowed or denied. Further, the machine learning techniques determine various other products that the customer may be interested in and prioritizes the choices of options that the machine learning algorithm determines appropriate for the customer.

Abstract: A neural network-based rendering technique increases temporal stability and image fidelity of low sample count path tracing by optimizing a distribution of samples for rendering each image in a sequence. A sample predictor neural network learns spatio-temporal sampling strategies such as placing more samples in dis-occluded regions and tracking specular highlights. Temporal feedback enables a denoiser neural network to boost the effective input sample count and increases temporal stability. The initial uniform sampling step typically present in adaptive sampling algorithms is not needed. The sample predictor and denoiser operate at interactive rates to achieve significantly improved image quality and temporal stability compared with conventional adaptive sampling techniques.

Abstract: A physical environment is equipped with a plurality of sensors (e.g., motion sensors). As individuals perform various activities within the physical environment, sensor readings are received from one or more of the sensors. Based on the sensor readings, activities being performed by the individuals are recognized and the sensor data is labeled based on the recognized activities. Future activity occurrences are predicted based on the labeled sensor data. Activity prompts may be generated and/or facility automation may be performed for one or more future activity occurrences.

Abstract: A method includes defining a first virtual being (e.g., including sensory locations for sensors, sense locations for sense properties, artificial neural networks connecting sensors to sense properties) in a virtual environment. The method also includes defining an object (e.g., including sense locations) in the virtual environment. The method also includes, in accordance with an interaction between the virtual being and the object, receiving sensory input at a first sensor at a first sensory location using a first virtual medium according to a first sense property of the object at a first sense location. The first sensor, the first virtual medium, and the first sense property have a same sensory type. According to the received sensory input, a first artificial neural network translates the received sensory input into updates to one or more configuration parameters of sensors of the first virtual being or movement of the virtual being.

Abstract: A fact validation method including the following steps: a statement to be validated is inputted and a searching is made for the statement to obtain an evidence set of the statement; a hierarchical heterogeneous graph consisting of an entity node, a sentence node and a context node is constructed based on the evidence set; the statement and the evidence set are spliced and a node is initialized to obtain feature representation of the node; the feature representation of the node is updated based on inference according to a propagation direction of a neural network of the node in the hierarchical heterogeneous graph; and an inference path for the updated feature representation of the node is built and a prediction result of the statement is output according to the inference path.

Abstract: Systems and methods analyze training of a first machine learning system with a second machine learning system. The first machine learning system comprises a neural network with a first inner layer node. The method includes connecting the first machine learning system to an input of the second machine learning system. The second machine learning system comprises a second objective function for analyzing an internal characteristic of the first machine learning system and which is different from a first objective function for the first machine learning system.

Abstract: Some embodiments of the invention provide a novel method for training a quantized machine-trained network. Some embodiments provide a method of scaling a feature map of a pre-trained floating-point neural network in order to match the range of output values provided by quantized activations in a quantized neural network. A quantization function is modified, in some embodiments, to be differentiable to fix the mismatch between the loss function computed in forward propagation and the loss gradient used in backward propagation. Variational information bottleneck, in some embodiments, is incorporated to train the network to be insensitive to multiplicative noise applied to each channel. In some embodiments, channels that finish training with large noise, for example, exceeding 100%, are pruned.

Abstract: A non-transitory computer-readable storage medium storing therein a learning program that causes a computer to execute a process includes: determining whether or not there is a discontinuity point at which a variation in a learning time relative to a variation in a learning parameter is discontinuous; specifying, when the discontinuity point is present, ranges of the learning parameter in which the variation in the learning time relative to the variation in the learning parameter is continuous, based on the discontinuity point; calculating, for each of the specified ranges, an estimated value of performance of trials using a trial parameter learned by machine learning per a learning time of machine learning using a learning parameter included in the range; and specifying a learning parameter which enables any of the estimated values selected in accordance with a magnitude of the estimated value among the calculated estimated values.

Abstract: Technologies are described for analyzing interactions with data objects stored by a network-based storage service. The analysis of the interactions can identify patterns of the data object interactions and outcomes that can result from the patterns. Models can be developed that include the patterns and the outcomes corresponding to the patterns. As requests related to data object interactions are subsequently obtained by the system, the requests can be analyzed with respect to the models to identify an outcome that may be associated with the requests.

Abstract: Techniques are disclosed for systems and methods for learning the behavior of and/or for performing automated testing of a system under test (SUT). The learning/testing is accomplished solely via the graphical user interface (GUI) of the SUT and requires no a priori metadata/knowledge about the GUI objects. The learning engine operates by performing actions on the GUI and by observing the results of these actions. If the actions result in a change in the screen/page of the GUI then a screenshot is taken for further processing. Objects are detected from the screenshot, new actions that need to be performed on the objects are guessed, those actions are performed, the results are observed and the process repeats. Text labels on the screens are also read and are used for generating contextualized inputs for the screens. The learning process continues until any predetermined learning/testing criteria are satisfied.

Abstract: An algorithm data store may contain information about a pool of available algorithms (e.g., to improve operation of an industrial asset). A deployment platform may be implemented in an edge portion at an industrial site associated with a live environment executing a current algorithm. A lifecycle manager of the deployment platform may manage execution of the current algorithm in the live environment creating source data. A performance manager may receive an indication of a selected at least one potential replacement algorithm from the pool of available algorithms and manage execution of the at least one potential replacement algorithm in a shadow environment using the source data. The performance manager may then report performance information associated with the at least one potential replacement algorithm. When appropriate, the potential replacement algorithm may replace the current algorithm.