Abstract: A method for enhancing kernel reparameterization of a non-linear machine learning model includes providing a predefined machine learning model, expanding a kernel of the predefined machine learning model with a non-linear network for convolution operation of the predefined machine learning model to generate the non-linear machine learning model, training the non-linear machine learning model, reparameterizing the non-linear network back to a kernel for convolution operation of the non-linear machine learning model to generate a reparameterized machine learning model, and deploying the reparameterized machine learning model to an edge device.

Abstract: In aspects of the disclosure, a method, a system, and a computer-readable medium are provided. The method of building a kernel reparameterization for replacing a convolution-wise operation kernel in training of a neural network comprises selecting one or more blocks from tensor blocks and operations; and connecting the selected one or more blocks with the selected operations to build the kernel reparameterization. The kernel reparameterization has a dimension same as that of the convolution-wise operation kernel.

Abstract: A method for removing branches from trained deep learning models is provided. The method includes steps (i)-(v). In step (i), a trained model is obtained. The trained model has a branch structure involving one or more original convolutional layers and a shortcut connection. In step (ii), the shortcut connection is removed from the branch structure. In step (iii), a reparameterization model is built by linearly expanding each of the original convolutional layers into a reparameterization block in the reparameterization model. In step (iv), parameters of the reparameterization blocks are optimized by training the reparameterization model. In step (v), each of the optimized reparameterization blocks is transformed into a reparameterized convolutional layer to form a branchless structure that replaces the branch structure in the trained model.

Abstract: The present disclosure provides an electronic module including a circuit including a transmitting part and a receiving part physically separated from the transmitting part. The electronic module also includes an element isolated from the circuit and configured to block electrical interference between the transmitting part and the receiving part.

Abstract: The invention introduces an apparatus for detecting errors during data encryption. The apparatus includes a search circuitry and a substitution check circuitry. The key generation circuitry is arranged operably to convert a first value of one byte corresponding to a plaintext, an intermediate encryption result, or a round key into a second value of a K-bit according to an 8-to-K lookup table, where K is an integer ranging from 10 to 15 and the second value comprises (K minus 8) bits of a Hamming parity. The substitution check circuitry is arranged operably to employ check formulae corresponding to the 8-to-K lookup table to determine whether an error is occurred during a conversion of the first value of the one byte into the second value of the K-bit, and output an error signal when finding the error, where a total amount of the formulae is K minus 8.

Abstract: The present invention provides a device for separating and purifying the collagen Type 2 in chicken bones, comprising a liquid fluid container, a raw liquid container, a membrane separation tank, a mixing tube, two high pressure metering motors, a precooler, two preheaters, a temperature controller, two one-way valves, two inlet control valves and two outlet control valves. The liquid extract of defatted chicken bones discharged from the raw liquid container and the liquid CO2 discharged from the liquid fluid container can be mixed uniformly in the mixing tube, and then fed into the membrane separation tank. The membrane separation tank produces small-molecular-weight peptides and large-molecular-weight collagen Type 2 harmlessly and efficiently.

Abstract: Systems and methods for improving seal performance in a centrifugal water pump for an internal combustion engine increase static pressure at the seal by incorporating a combination of slots and ribs into a seal cavity of the pump housing that are positioned to convert dynamic fluid pressure into static pressure at the seal while reducing coolant velocity at the seal. Vent holes in the impeller having an appropriate size and location may also be used to increase the static pressure at the seal to enhance seal performance.