Abstract: A load sensor compares generated values of load signals to predicted values of the load signals. Where an error between a generated value and a predicted value is less than a threshold, a value of zero is transmitted by the load sensor to an external systems, in an encoded stream of bits, and the error are used to calculate a predicated value of a next load signal, such that insubstantial changes in loading resulting in small errors are not transmitted to the external system. Where an error between the generated value and the predicted value exceeds the threshold, however, a value representative of the error is transmitted to the external system, in an encoded stream of bits. The external system processes the stream of bits to calculate the error, to determine that an item has been placed on a storage system or removed therefrom, or to identify the item.

Abstract: A distributed deep neural net (DNN) utilizing a distributed, tile-based architecture implemented on a semiconductor package. The package includes multiple chips, each with a central processing element, a global memory buffer, and processing elements. Each processing element includes a weight buffer, an activation buffer, and multiply-accumulate units to combine, in parallel, the weight values and the activation values.

Abstract: Quantizing tensors and vectors processed within a neural network reduces power consumption and may accelerate processing. Quantization reduces the number of bits used to represent a value, where decreasing the number of bits used can decrease the accuracy of computations that use the value. Ideally, quantization is performed without reducing accuracy. Quantization-aware training (QAT) is performed by dynamically quantizing tensors (weights and activations) using optimal clipping scalars. “Optimal” in that the mean squared error (MSE) of the quantized operation is minimized and the clipping scalars define the degree or amount of quantization for various tensors of the operation. Conventional techniques that quantize tensors during training suffer from high amounts of noise (error). Other techniques compute the clipping scalars offline through a brute force search to provide high accuracy.

Abstract: An electronic lock includes an interior subassembly including a manual turnpiece, a control circuit, and a motor actuatable by the control circuit, and an exterior subassembly including a facial authentication circuit operatively connected to the control circuit. The electronic lock includes a latch operatively connected to the interior subassembly, the latch being engageable by the manual turnpiece or the motor to move the electronic lock between a locked state and an unlocked state. The facial authentication circuit includes at least one camera and is configured to generate an authentication signal to the control circuit, and wherein, at least in part in response to the authentication signal, the control circuit is configured to actuate the motor to move the electronic lock between the locked state and the unlocked state.

Abstract: A polarizing beam splitter assembly for directing image light on an input path into multiple exit light paths comprises multiple prisms with edges that meet to form a seam. The polarizing beam splitter assembly includes a diffracting element prior to the seam in the input light path. The diffracting element comprises a geometry that performs at least one of blocking a portion of the image light and scattering a portion of the image light.

Abstract: An error reporting system utilizes a parity checker to receive data results from execution of an original instruction and a parity bit for the data. A decoder receives an error correcting code (ECC) for data resulting from execution of a shadow instruction of the original instruction, and data error correction is initiated on the original instruction result on condition of a mismatch between the parity bit and the original instruction result, and the decoder asserting a correctable error in the original instruction result.

Abstract: Described is the efficient and robust generation and isolation of functional astrocytes from pluripotent stem cells (PSCs). The methodology provided recapitulate the major steps of oligodendrocyte differentiation into mixed cell populations and subsequent isolation of astrocytes from the mixed cell populations.

Abstract: Video routing may include integrated audio mixing, audio processing, or both. An audio subsystem that is integrated with a video router, for example, may receive and mix router input audio signals to provide mixed audio signals, and route the mixed audio signals to router outputs as router output audio signals. Regarding audio processing, input audio signals that include router input audio signals that are received by the video router may be routed as output audio signals. The output audio signals include router output audio signals to be output from the video router. Respective ones of the output audio signals are processed to provide respective processed output audio signals. The input audio signals for the routing also include the processed output audio signals.

Abstract: A polarizing beam splitter assembly for directing image light on an input path into multiple exit light paths comprises multiple prisms with edges that meet to form a seam. The polarizing beam splitter assembly includes a diffracting element prior to the seam in the input light path. The diffracting element comprises a geometry that performs at least one of blocking a portion of the image light and scattering a portion of the image light.

Abstract: A polarizing beam splitter assembly for directing image light on an input path into multiple exit light paths comprises multiple prisms with edges that meet to form a seam. The polarizing beam splitter assembly includes a diffracting element prior to the seam in the input light path. The diffracting element comprises a geometry that performs at least one of blocking a portion of the image light and scattering a portion of the image light.

Abstract: A polarizing beam splitter assembly for directing image light on an input path into multiple exit light paths comprises multiple prisms with edges that meet to form a seam. The polarizing beam splitter assembly includes a diffracting element prior to the seam in the input light path. The diffracting element comprises a geometry that performs at least one of blocking a portion of the image light and scattering a portion of the image light.

Abstract: A distributed deep neural net (DNN) utilizing a distributed, tile-based architecture implemented on a semiconductor package. The package includes multiple chips, each with a central processing element, a global memory buffer, and processing elements. Each processing element includes a weight buffer, an activation buffer, and multiply-accumulate units to combine, in parallel, the weight values and the activation values.

Abstract: A single-ended signal transmission system recovers a noise signal associated with a data input signal and uses the recovered noise signal to compensate for noise on the data input signal. The noise signal may be recovered from a noise reference signal line, or clock signal line, or a data signal line associated with a DC-balanced data input signal. The recovered noise signal may be represented as an analog signal or a digital signal. The recovered noise signal may be processed to compensate for DC offset and nonlinearities associated with one or more different input buffers. In one embodiment, the recovered noise signal includes frequency content substantially below a fundamental frequency for data transmission through the data input signal.

Abstract: A single-ended signal transmission system recovers a noise signal associated with a data input signal and uses the recovered noise signal to compensate for noise on the data input signal. The noise signal may be recovered from a noise reference signal line, or clock signal line, or a data signal line associated with a DC-balanced data input signal. The recovered noise signal may be represented as an analog signal or a digital signal. The recovered noise signal may be processed to compensate for DC offset and nonlinearities associated with one or more different input buffers. In one embodiment, the recovered noise signal includes frequency content substantially below a fundamental frequency for data transmission through the data input signal.

Abstract: A single-ended signal transmission system recovers a noise signal associated with a data input signal and uses the recovered noise signal to compensate for noise on the data input signal. The noise signal may be recovered from a noise reference signal line, or clock signal line, or a data signal line associated with a DC-balanced data input signal. The recovered noise signal may be represented as an analog signal or a digital signal. The recovered noise signal may be processed to compensate for DC offset and nonlinearities associated with one or more different input buffers. In one embodiment, the recovered noise signal includes frequency content substantially below a fundamental frequency for data transmission through the data input signal.

Abstract: A single-ended signal transmission system recovers a noise signal associated with a data input signal and uses the recovered noise signal to compensate for noise on the data input signal. The noise signal may be recovered from a noise reference signal line, or clock signal line, or a data signal line associated with a DC-balanced data input signal. The recovered noise signal may be represented as an analog signal or a digital signal. The recovered noise signal may be processed to compensate for DC offset and nonlinearities associated with one or more different input buffers. In one embodiment, the recovered noise signal includes frequency content substantially below a fundamental frequency for data transmission through the data input signal.

Abstract: Computer-implemented methods for adding story content to a web site are provided. In one aspect, a computer-implemented method provides for visitors to a website or single-page application (SPA) to add content to a website or SPA based on a storytelling rule set. In another aspect, a computer-implemented method provides for visitors to a website or SPA to add content of a type within a frame of the same type.

Abstract: A processor and a system are provided for tuning a supply voltage for data retention. The contents of data storage circuitry are read and a data verification indication corresponding to the contents is computed. Then, the supply voltage provided to the data storage circuitry is reduced to a low voltage level that is intended to retain the contents of the data storage circuitry.

Abstract: A method and a system are provided for performing write assist. Write assist circuitry is initialized and voltage collapse is initiated to reduce a column supply voltage provided to a storage cell. A bitline of the storage cell is boosted to a boosted voltage level that is below a low supply voltage provided to the storage cell and data encoded by the bitline is written to the storage cell.

Abstract: A method and a system are provided for performing write assist. Write assist circuitry is initialized and voltage collapse is initiated to reduce a column supply voltage provided to a storage cell. A bitline of the storage cell is boosted to a boosted voltage level that is below a low supply voltage provided to the storage cell and data encoded by the bitline is written to the storage cell.