Abstract: Methods, systems and computer readable media using hardware-efficient bit-shift operations for computing the output of a low-bit neural network layer. A dense shift inner product operator (or dense shift IPO) using bit shifting in place of multiplication replaces the inner product operator that is conventionally used to compute the output of a neural network layer. Dense shift neural networks may have weights encoded using a low-bit dense shift encoding. A dedicated neural network accelerator is designed to compute the output of a dense shift neural network layer using dense shift IPOs. A Sign-Sparse-Shift (S3) training technique trains a low-bit neural network using dense shift IPOs or other bit shift operations in computing its outputs.

Abstract: Disclosed is a high-temperature virus-killing mask, including a sunhat, a heating component and a mask body. The heating component comprises a high-temperature microtube and a base, both arranged in the sunhat. The high-temperature microtube is provide with a resistance wire which may be energized to generate heat to enable the temperature in the high-temperature microtube to rise up to about 200° C.

Abstract: Disclosed is an infrared thermal image classification and hot spot positioning method of a photovoltaic panel, comprising following steps: constructing a photovoltaic panel infrared thermal image data set, preprocessing the image data set, and dividing into a training set and a testing set according to a preset proportion; training an auxiliary generating countermeasure network based on the training set to obtain a trained generator and a trained discriminator, training an encoder by combining the image data in the training set with the generator, and fixing the parameters of a trained encoder; inputting the image data in the test set into a trained discriminator to obtain a photovoltaic image classification result; inputting images in the test set into the trained encoder, and then inputting the images into the generator for a reconstruction; comparing pixels of the input images with the reconstructed infrared thermal images, and positioning hot spots.

Abstract: Systems and methods for computing a neural network layer of a neural network are described. A squared Euclidean distance is computed between the input vector and the weight vector of the neural network layer, replacing computation of the inner product. Methods for quantization of the squared Euclidean computation are also described. Methods for training the neural network using homotopy training are also described.

Abstract: Disclosed is a waterseal virus-killing mask, including a mask body, a liquid virus-killing assembly, and two elastic ear loops. A breather valve is disposed on an outer surface of the mask body. The two elastic ear loops are disposed on two sides of the mask body, respectively. The liquid virus-killing assembly includes a U-shaped tube within which disinfectant is provided. A disinfectant replenishing tube is vertically disposed at an upper end in the middle of the U-shaped tube. A first horizontal straight tube and a second horizontal straight tube are separately disposed at two ends of the U-shaped tube. A left end of the first horizontal straight tube is in communication with an opening of the breather valve. A first filter mesh used to separate water and air is disposed at a right end of the first horizontal straight tube.

Abstract: A computational block configured to perform an inference task by applying a plurality of low resource computing operations to a binary input feature tensor to generate an integer feature tensor that is equivalent to an output of multiplication and accumulation operations performed in respect of a ternary weight tensor and the binary input feature tensor; and performing a comparison operation between the generated integer feature tensor and a comparison threshold to generate a binary output feature tensor.

Abstract: Disclosed is a waterseal virus-killing mask, including a mask body, a liquid virus-killing assembly, and two elastic ear loops. A breather valve is disposed on an outer surface of the mask body. The two elastic ear loops are disposed on two sides of the mask body, respectively. The liquid virus-killing assembly includes a U-shaped tube within which disinfectant is provided. A disinfectant replenishing tube is vertically disposed at an upper end in the middle of the U-shaped tube. A first horizontal straight tube and a second horizontal straight tube are separately disposed at two ends of the U-shaped tube. A left end of the first horizontal straight tube is in communication with an opening of the breather valve. A first filter mesh used to separate water and air is disposed at a right end of the first horizontal straight tube.

Abstract: Disclosed is a high-temperature virus-killing mask, including a sunhat, a heating component and a mask body. A high-temperature microtube with a resistance wire is arranged in the sunhat. The resistance wire may be energized to generate heat to enable the temperature in the high-temperature microtube to rise up to about 200° C., so that, when a user inhales air, the NCP in air flowing through the high-temperature microtube will be killed by the high temperature.

Abstract: A method of training a neural network (NN) block for a neural network, including: performing a first quantization operation on a real-valued feature map tensor to generate a corresponding binary feature map tensor; performing a second quantization operation on a real-valued weight tensor to generate a corresponding binary weight tensor; convoluting the binary feature map tensor with the binary weight tensor to generate a convoluted output; scaling the convoluted output with a scaling factor to generate a scaled output, wherein the scaled output is equal to an estimated weight tensor convoluted with the binary feature map tensor, the estimated weight tensor corresponding to a product of the binary weight tensor and the scaling factor; calculating a loss function, the loss function including a regularization function configured to train the scaling factor so that the estimated weight tensor is guided towards the real-valued weight tensor; and updating the real-valued weight tensor and scaling factor based on th