Abstract: Sample classification using natural language processing (NLP) models is disclosed herein. An example apparatus comprises interface circuitry, machine readable instructions, and programmable circuitry to at least one of instantiate or execute the machine readable instructions to access a set of instructions, the set of instructions representing executable instructions, determine an Application Programming Interface (API) call sequence based on the set of instructions, transmit the API call sequence to a NLP model, the NLP model to generate a set of tokens, the set of tokens representing the API call sequence, the set of tokens readable in the NLP model, and classify the API call sequence as clean or malicious based on the tokens, and classify the set of instructions as clean or malicious based on the classification of the API call sequence and at least one other feature of the set of instructions.

Abstract: Systems and methods for identifying substitute elements for computer-implemented processes are disclosed. A substitution request identifying an anchor element is received. A set of candidate substitution elements is generated by a trained candidate selection model configured to receive the anchor element, a feature set, and a set of catalog elements. The set of candidate substitution elements is ranked by a trained ranking model configured to receive the anchor element, the feature set, and the set of candidate substitution elements. At least one substitution element is selected from the set of candidate substitution elements and feedback data representative of the suitability of the selected at least one substitution element with respect to the anchor element is received. At least one of the trained candidate selection model or the trained ranking model is updated by applying an iterative training process incorporating at least a portion of the feedback data.

Abstract: Methods, systems, and computer program products are provided for coordinated analysis of output scores and input features of machine learning models in different environments. An example method includes receiving a plurality of first data records and a plurality of second data records. A first plot is generated based on a first score generated by a machine learning model for each first data record and a second score generated by the machine learning model for each second data record. The first plot is displayed. A plurality of second plots associated with at least a subset of the plurality of features are generated. Each respective second plot is generated based on a respective first field associated with a respective feature from the first data records and a respective second field associated with the respective feature from the second data records. The second plots are displayed.

Abstract: Provided is a system for analyzing features associated with entities using an embedding tree, the system including at least one processor programmed or configured to receive a dataset associated with a plurality of entities, wherein the dataset comprises a plurality of data instances for a plurality of entities. The processor may be programmed or configured to generate at least two embeddings based on the dataset and determine split criteria for partitioning an embedding space of at least one embedding tree associated with the dataset based on feature data associated with an entity and embedding data associated with the at least two embeddings. The processor may be programmed or configured to generate at least one embedding tree having a plurality of nodes based on the split criteria. Methods and computer program products are also provided.

Abstract: Systems, apparatuses, and methods for an on chip dynamic IR oscilloscope are provided. An oscilloscope circuitry may comprise sensor circuitry, voltage generator circuitry, finite state machine, and latch circuitry. The sensor circuitry may include digital logic circuitry, sample and hold circuitry, and sense amplifier circuitry. The voltage generator circuitry may include a voltage generator, analog buffers, switches, and high speed buffer. The finite state machine may control the sensor circuitry to sample a voltage waveform and the voltage generator circuitry to generate a reference voltage that may change over time. The sensing amplifier circuitry may compare the samples to the reference voltage to generate flags when a sample exceeds a reference voltage. The flags may be used to stored the voltages associated with the flags, which may be used to redraw the waveform sampled.

Abstract: A device for graphical rendering includes a memory and processing circuitry. The processing circuitry is configured to receive sample values, transmitted by one or more servers, of samples of an object, wherein the sample values are generated by the one or more servers from inputting coordinates into a trained neural network and outputting, from the trained neural network, the sample values of the samples, store the sample values in the memory, and render image content of the object based on the sample values.

Abstract: A method for securely sharing of data by an electronic device is provided. The method includes receiving, by the electronic device, data associated with the at least one application available at the electronic device and obtaining, by the electronic device, secured data by transforming the data associated with at least one application into an unrecognizable format using at least one conceal factor and at least one noise input. Further, the method includes extracting, by the electronic device, a plurality of features from the secured data, and sharing, by the electronic device, the plurality of features extracted from the secured data to a plurality of servers.

Abstract: Systems, methods, and computer program products that obtain a plurality of features associated with a plurality of samples and a plurality of labels for the plurality of samples; generate a plurality of first predictions for the plurality of samples with a first machine learning model; generate a plurality of second predictions for the plurality of samples with a second machine learning model; generate, based on the plurality of first predictions, the plurality of second predictions, the plurality of labels, and a plurality of groups of samples of the plurality of samples; determine, based on the plurality of groups of samples, a first success rate associated with the first machine learning model and a second success rate associated with the second machine learning model; and identify, based on the first success rate and the second success rate, a weak point in the machine learning first model or the second model.

Abstract: A system for efficient element-wise and cross-vector maximum operations. One example system includes an input bus, an output bus, and a memory configured to store N/K elements of an N-element vector in a corresponding row of the memory. A K-wide FMAX( ) comparator has a first set of K-wide inputs of the K-wide FMAX( ) comparator coupled to a read port of the memory and a second set of K-wide inputs of the K-wide FMAX( ) comparator coupled to the input bus, and a set of K-wide outputs of the K-wide FMAX( ) comparator coupled to a write port of the memory. A tree of FMAX( ) comparators comprises an input and an output, the input of the tree coupled to the set of K-wide outputs of the K-wide FMAX( ) comparator and the output of the tree coupled to the output bus.

Abstract: Disclosed are example embodiments of systems and methods for generating one or more recommendations based on, at least in part, unstructured data. In an example embodiment, a method generally includes accessing transaction data representative of a plurality of transactions, the transaction data including unstructured data describing ones of the plurality of transactions and structured data indicative of the transactions, where at least a portion of the transactions involve a user. The method also includes compiling, by a computing device, using a convolution neural network (CNN) with the unstructured data from the transaction data and a recurrent neural network (RNN) with structure data of the transaction data, one or more feature vector indicative of the user. The method then includes generating a recommendation based on the feature vector for the user and publishing the recommendation.

Abstract: The present disclosure relates to a method and apparatus for generating training data of a deep learning model for lane classification. The method according to an embodiment of the present disclosure is performed by an electronic apparatus, and is a method for generating training data of a deep learning model for lane classification by generating a composite image of the other color lane using images of a white lane and the other color lane, and includes determining a ratio of other two channels based on one channel (reference color channel) for three color channels of red (R), green (G) and blue (B) of the other color lane in the image of the other color lane; and generating a composite image of the other color lane by scaling the image of the white lane by applying the determined ratio to the other two channels with respect to the reference color channel of the white lane.

Abstract: A device for determining lane type and method thereof are provided. The device for determining lane type according to an embodiment of the present disclosure includes a camera for acquiring an around view image around a vehicle, a GPS receiver for receiving GPS information, and a controller communicatively connected to the camera and the GPS receiver. Here, the controller is configured to recognize a scene of the image acquired by the camera, detect lanes and road markings from the recognized scene, comprise a classifier adapted based on the GPS information, classify the detected lanes and road markings by the classifier, and confirm a type of the classified lanes.

Abstract: The present disclosure is related in general to field of machine learning and image processing, that provides a method and system for detecting and classifying lanes. A lane detection and classification system receives an input image from a data source and segments the input image into plurality of segments using a trained semantic segmentation model. Further, one or more lane markings are detected in the segmented image and, lane pattern and lane colour of each of the one or more lane markings, wherein each lane marking is associated with a priority. Subsequently, a binary image comprising lane markings of ego lanes of the host vehicle is generated and coefficient values of the ego lanes of the host vehicle are determined based on the priority value associated with the lane markings of the ego lanes and current position of the host vehicle, using a trained Convolutional Neural Network (CNN) model.

Abstract: A compensated deep neural network (compensated-DNN) is provided. A first vector having a set of components and a second vector having a set of corresponding components are received. A component of the first vector includes a first quantized value and a first compensation instruction, and a corresponding component of the second vector includes a second quantized value and a second compensation instruction. The first quantized value is multiplied with the second quantized value to compute a raw product value. The raw product value is compensated for a quantization error according to the first and second compensation instructions to produce a compensated product value. The compensated product value is added into an accumulated value for the dot product. The accumulated value is converted into an output vector of the dot product. The output vector includes an output quantized value and an output compensation instruction.

Abstract: An efficient pipelined implementation of digital scaling, offset and aggregation operation supports element-by-element programmable scale and offset factors. The method includes time-multiplexed parallel pipelining of a plurality of digital data words, each of the plurality of digital data words encoding an N-bit signed integer, from one of a plurality of receive-registers through a datapath that can either (1) store the plurality of digital data words directly in a dedicated first memory, (2) store the plurality of digital data words directly in a dedicated second memory, or (3) direct the plurality of digital data words into a parallel set of fused-multiply-add units. The method further includes multiplying each digital data word by a corresponding data-word retrieved from the dedicated first memory to form product data words and adding the product data words to a corresponding data-word retrieved from the dedicated second memory to form an output sum-and-product data words.

Abstract: A compute engine is configured to perform self-attention computations by delaying performance of a division operation of a softmax computation, the performance including iteratively computing a first matrix multiplication of a given row vector of a first matrix and each column vector of a second matrix while determining a first scalar element representing a maximum value of the iterative first matrix multiplications; iteratively subtracting a corresponding determined first scaler element from a result of each computed first matrix multiplication and computing an elementwise exponential function based on a result of the subtraction operation to generate a plurality of elements of a given row vector of a fourth matrix; iteratively computing a second matrix multiplication of a given row vector of the fourth matrix and each column vector of a third matrix while summing the given row vectors of the fourth matrix; and computing a row vector of an output matrix.

Abstract: Systems and methods for performing layer normalization are described. A circuit can receive a sequence of input data across a plurality of clock cycles, where the sequence of input data represents a portion of an input vector. The circuit can determine a plurality of sums and a plurality of sums of squares corresponding to the sequence of input data. The circuit can determine, based on the plurality of sums of squares, a first scalar representing an inverse square-root of a variance of vector elements in the input vector. The circuit can determine a second scalar representing a negation of a product of the first scalar and a mean of the vector elements in the input vector. The circuit can determine, based on the first scalar, the second scalar and the received sequence of input data, an output vector that is a normalization of the input vector.

Abstract: A method for improving consensus in a blockchain network through decentralized grouping includes: identifying, by each node of a plurality of nodes in a blockchain network that manages a blockchain, a plurality of groups, where each is comprised of a subset of nodes; generating, by each node in each subset of nodes, a new block for the blockchain; performing, by each subset of nodes, a first consensus operation among all nodes in the subset of nodes for the new block generated by in the subset of nodes to identify a group consensus block, where each node in the subset of nodes receives the group's group consensus block; and performing, by the blockchain network, a second consensus operation among all groups for the identified group consensus block to identify an overall consensus block, where a majority of groups of the plurality of groups receives the overall consensus block.

Abstract: Systems, methods, and computer program products may compare machine learning models by identifying data instances with disagreed predictions and learning from the disagreement. Based on a model interpretation technique, differences between the compared machine learning models may be interpreted. Multiple metrics to prioritize meta-features from different perspectives may also be provided.

Abstract: Provided is a system for analyzing features associated with entities using an embedding tree, the system including at least one processor programmed or configured to receive a dataset associated with a plurality of entities, wherein the dataset comprises a plurality of data instances for a plurality of entities. The processor may be programmed or configured to generate at least two embeddings based on the dataset and determine split criteria for partitioning an embedding space of at least one embedding tree associated with the dataset based on feature data associated with an entity and embedding data associated with the at least two embeddings. The processor may be programmed or configured to generate at least one embedding tree having a plurality of nodes based on the split criteria. Methods and computer program products are also provided.