Abstract: Various embodiments provide for deep learning-based architectures and design methodologies for an orthogonal frequency division multiplexing (OFDM) receiver under the constraint of one-bit complex quantization. Single bit quantization greatly reduces complexity and power consumption in the receivers, but makes accurate channel estimation and data detection difficult. This is particularly true for OFDM waveforms, which have high peak-to average (signal power) ratio in the time domain and fragile subcarrier orthogonality in the frequency domain. The severe distortion for one-bit quantization typically results in an error floor even at moderately low signal-to-noise-ratio (SNR) such as 5 dB. For channel estimation (using pilots), various embodiments use novel generative supervised deep neural networks (DNNs) that can be trained with a reasonable number of pilots. After channel estimation, a neural network-based receiver specifically, an autoencoder jointly learns a precoder and decoder for data symbol detection.

Abstract: Embodiments of the disclosure are drawn to implantable stimulator with machine learning based classifier. An implantable system includes sensors which provide sensor information to an implantable unit. The implantable unit uses a classifier on the sensor information to select a stimulation procedure which is applied via a stimulation electrode. The classifier may be generated by a trained machine learning model. The classifier may be trained on an external unit which is not implanted in the subject. The classifier may be trained based on sensor information from the implanted sensors as well as symptom information.

Abstract: Techniques are described for managing data transfer between an enterprise system and a third-party provider system, through a batch interface of the third-party provider system. A batch of transaction data is uploaded through the batch interface for enrichment, and enrichment data is received in response. The enrichment data provides additional context regarding each of the transactions specified in the uploaded batch. The enrichment data can be sent back to the presented to user(s) through various tools. Prior to uploading, the batch can be filtered to include those records corresponding to users who have opted into a service. Filtering can also exclude records for which enrichment data has already been provided through a real-time process. Reconciliation can be performed to ensure integrity of the communicated data.

Abstract: An improved internal combustion engine air intake system that operates in tandem with the original equipment air intake system and does not endanger the vehicle warranty and provides improved airflow to the engine which improves engine performance and efficiency. The improved internal combustion engine air intake system also provides one or more additional air inputs taken from a different location than the original equipment air intake which improves the likelihood of providing additional cooler, denser, air to the engine air intake which will improve engine efficiency.

Abstract: Embodiments of the present disclosure include systems and methods for accelerating processing based on sparsity for neural network hardware processors. An input manager determines a pair of non-zero values from a pair of data streams in a plurality of pairs of data streams and retrieve the pair of non-zero values from the pair of data streams. A multiplier performs a multiplication operation on the pair of non-zero values and generate a product of the pair of non-zero values. An accumulator manager receives the product of the pair of non-zero values from the multiplier and sends the product of the pair of non-zero values to a corresponding accumulator in a plurality of accumulators.

Abstract: Embodiments of the present disclosure include systems and methods for determining schedules for processing neural networks on hardware. A set of instructions for processing data through a neural network is received. Based on a hardware definition specifying the set of hardware units and functions that each hardware unit in the set of the hardware unit is configured to perform, a schedule of a set of operations to be performed by a subset of the set of hardware units to implement the set of instructions are determined. The schedule of the set of operations are distributed to the subset of the set of hardware units.

Abstract: Various embodiments provide for deep learning-based architectures and design methodologies for an orthogonal frequency division multiplexing (OFDM) receiver under the constraint of one-bit complex quantization. Single bit quantization greatly reduces complexity and power consumption in the receivers, but makes accurate channel estimation and data detection difficult. This is particularly true for OFDM waveforms, which have high peak-to average (signal power) ratio in the time domain and fragile subcarrier orthogonality in the frequency domain. The severe distortion for one-bit quantization typically results in an error floor even at moderately low signal-to-noise-ratio (SNR) such as 5 dB. For channel estimation (using pilots), various embodiments use novel generative supervised deep neural networks (DNNs) that can be trained with a reasonable number of pilots. After channel estimation, a neural network-based receiver specifically, an autoencoder jointly learns a precoder and decoder for data symbol detection.

Abstract: Various embodiments provide for testing a transmitter using a phase-lock loop, which can be used with a circuit for data communications, such as serializer/deserializer (SerDes) communications. In particular, some embodiments provide for data transmission test of a transmitter by: generating and outputting a pre-determined data pattern through a serializer of the transmitter; sampling a serialized data output of the serializer using a sample clock signal generated by an M/N phase-lock loop (PLL); and using a pattern checker to error check the sampled data to determine whether the data transmission test passes.

Abstract: Methods, computer systems, and computer storage media are provided for enabling the reuse of data structures, data processing, data cleaning, data mapping, analytic concepts, analytic widgets, logic, and graphical components to efficiently provide custom insights to new areas. Additionally, the reuse of data structures, data processing, data cleaning, data mapping, analytic concepts, analytic widgets, logic, and graphical components created for a single model can also be reused components across new models, which enables new models to be efficiently created for a completely separate set of analytics. A model is inimitably created by creating, cleaning, and/or mapping a data concept and creating widgets that provide analytic capabilities for the model. These data concepts and widgets are specifically designed to be reused across models and model instances for a plurality of clients or users. Any data concepts and widgets of the model can also be modified to provide a customized solution.

Abstract: A motor is disclosed. The motor includes a first end bell, a second end bell and a stator with a stator coil disposed between the first end bell and the second end bell. A rotor with a rotor shaft is disposed relative to the stator, the rotor configured to rotate relative to the stator and the rotor shaft extending through the first end bell. The second end bell includes a chamber, the chamber includes an electronic circuit and a connector. The connector is electrically coupled to the electronic circuit and configured to receive both a control signal and a power signal from an external source.

Abstract: A system for use with a connection comprising a terminus opening into an underground vault. The system includes a perturbation detector and a system controller. The perturbation detector is configured to detect values of each of one or more properties and transmit a signal encoding the values to the system controller. The system controller is configured to receive the signal, use at least one of the values to determine whether a fire is occurring or has occurred, and alert a user when the system controller determines that the fire is occurring or has occurred.