Abstract: System and methods for compressing image-to-image models. Generative Adversarial Networks (GANs) have achieved success in generating high-fidelity images. An image compression system and method adds a novel variant to class-dependent parameters (CLADE), referred to as CLADE-Avg, which recovers the image quality without introducing extra computational cost. An extra layer of average smoothing is performed between the parameter and normalization layers. Compared to CLADE, this image compression system and method smooths abrupt boundaries, and introduces more possible values for the scaling and shift. In addition, the kernel size for the average smoothing can be selected as a hyperparameter, such as a 3×3 kernel size. This method does not introduce extra multiplications but only addition, and thus does not introduce much computational overhead, as the division can be absorbed into the parameters after training.

Abstract: Described is a system for improving machine learning models by accessing a first latent diffusion machine learning model, accessing a second latent diffusion machine learning model that was derived from the first latent diffusion machine learning model, the second latent diffusion machine learning model trained to perform a second number of denoising steps, generating noise data, processing the noise data via the first latent diffusion machine learning model to generate one or more first latent features, processing the noise data via the second latent diffusion machine learning model to generate one or more second latent features, and inputting the one or more first latent features and the one or more second latent features into a loss function. The system then modifies a parameter of the second latent diffusion machine learning model based on the output of the loss function.

Abstract: Described is a system for improving machine learning models. In some cases, the system improves such models by identifying a performance characteristic for machine learning model blocks in an iterative denoising process of a machine learning model, connecting a prior machine learning model block with a subsequent machine learning model block of the machine learning model blocks within the machine learning model based on the identified performance characteristic, identifying a prompt of a user, the prompt indicative of an intent of the user for generative images, and analyzing data corresponding to the prompt using the machine learning model to generate one or more images, the machine learning model trained to generate images based on data corresponding to prompts.

Abstract: Described is a system for improving machine learning models. In some cases, the system improves such models by identifying an autoencoder for a latent diffusion machine learning model, the latent diffusion machine learning model is trained to receive text as input and output an image based on the received text. The system identifies a number of channels in a decoder of the autoencoder, the decoder being configured to receive latent features as input and output images. The system further identifies a performance characteristic of the decoder and changes the node topology of the decoder based on the performance characteristic to generate an updated decoder. The system retrains the latent diffusion machine learning model using the updated decoder by inputting latent features to the updated decoder, receiving an outputted image from the updated decoder, and updating one or more weights of the decoder based on an assessment of the outputted image.

Abstract: Described is a system for improving machine learning models by accessing a first latent diffusion machine learning model, the first latent diffusion machine learning model trained to perform a first number of denoising steps, accessing a second latent diffusion machine learning model that was derived from the first latent diffusion machine learning model, the second latent diffusion machine learning model trained to perform a second number of denoising steps, generating noise data, processing the noise data via the first latent diffusion machine learning model to generate one or more first images, processing the noise data via the second latent diffusion machine learning model to generate one or more second images, and modify a parameter of the second latent diffusion machine learning model based on a comparison of the one or more first images with the one or more second images.

Abstract: System and methods for compressing image-to-image models. Generative Adversarial Networks (GANs) have achieved success in generating high-fidelity images. An image compression system and method adds a novel variant to class-dependent parameters (CLADE), referred to as CLADE-Avg, which recovers the image quality without introducing extra computational cost. An extra layer of average smoothing is performed between the parameter and normalization layers. Compared to CLADE, this image compression system and method smooths abrupt boundaries, and introduces more possible values for the scaling and shift. In addition, the kernel size for the average smoothing can be selected as a hyperparameter, such as a 3×3 kernel size. This method does not introduce extra multiplications but only addition, and thus does not introduce much computational overhead, as the division can be absorbed into the parameters after training.

Abstract: An adjustment method of moisture content and dense state for a hydrogel improved subgrade based on weather-resistance during an in-service period, including: step 1: carrying out surface cleaning and compaction of ground; step 2: preparing hydrogel improved subgrade raw material; step 3: paving the prepared material on the surface to form first-layer improved subgrade; and step 4: paving plain soil subgrade onto the first-layer. The method combines the water absorption and release function of the modified resin and the characteristic of the gel state thereof, to pave the improved subgrade in layers, which can absorb water and slightly expand when the water content in the subgrade is increased to a certain threshold value, and a strength and compactness protective layer can also be formed at the connection sections of the ground and the subgrade, and the subgrade and the pavement, to prevent the pot-cover effect from occurring.

Abstract: System and methods for compressing image-to-image models. Generative Adversarial Networks (GANs) have achieved success in generating high-fidelity images. An image compression system and method adds a novel variant to class-dependent parameters (CLADE), referred to as CLADE-Avg, which recovers the image quality without introducing extra computational cost. An extra layer of average smoothing is performed between the parameter and normalization layers. Compared to CLADE, this image compression system and method smooths abrupt boundaries, and introduces more possible values for the scaling and shift. In addition, the kernel size for the average smoothing can be selected as a hyperparameter, such as a 3×3 kernel size. This method does not introduce extra multiplications but only addition, and thus does not introduce much computational overhead, as the division can be absorbed into the parameters after training.

Abstract: System and methods for compressing image-to-image models. Generative Adversarial Networks (GANs) have achieved success in generating high-fidelity images. An image compression system and method adds a novel variant to class-dependent parameters (CLADE), referred to as CLADE-Avg, which recovers the image quality without introducing extra computational cost. An extra layer of average smoothing is performed between the parameter and normalization layers. Compared to CLADE, this image compression system and method smooths abrupt boundaries, and introduces more possible values for the scaling and shift. In addition, the kernel size for the average smoothing can be selected as a hyperparameter, such as a 3×3 kernel size. This method does not introduce extra multiplications but only addition, and thus does not introduce much computational overhead, as the division can be absorbed into the parameters after training.

Abstract: In a job management environment comprising a plurality of job systems and a scheduler for scheduling a job submitted to the job management environment to a job system for running, a processor, in a first job system, intercepts, from outside of a first container in the first job system, a first job from being sent to the scheduler. A processor, in the first job system, determines whether the first job is submitted from a container in the first job system. In response to a first determination that the first job is submitted from a container in the first job system, a processor, in the first job system, determines contexts of the first job, the contexts of the first job including a context related to the first container. A processor, in the first job system, sends the first job together with the contexts of the first job to the scheduler.

Abstract: Techniques for training a neural network having a plurality of computational layers with associated weights and activations for computational layers in fixed-point formats include determining an optimal fractional length for weights and activations for the computational layers; training a learned clipping-level with fixed-point quantization using a PACT process for the computational layers; and quantizing on effective weights that fuses a weight of a convolution layer with a weight and running variance from a batch normalization layer. A fractional length for weights of the computational layers is determined from current values of weights using the determined optimal fractional length for the weights of the computational layers. A fixed-point activation between adjacent computational layers is related using PACT quantization of the clipping-level and an activation fractional length from a node in a following computational layer.

Abstract: System and methods for compressing image-to-image models. Generative Adversarial Networks (GANs) have achieved success in generating high-fidelity images. An image compression system and method adds a novel variant to class-dependent parameters (CLADE), referred to as CLADE-Avg, which recovers the image quality without introducing extra computational cost. An extra layer of average smoothing is performed between the parameter and normalization layers. Compared to CLADE, this image compression system and method smooths abrupt boundaries, and introduces more possible values for the scaling and shift. In addition, the kernel size for the average smoothing can be selected as a hyperparameter, such as a 3×3 kernel size. This method does not introduce extra multiplications but only addition, and thus does not introduce much computational overhead, as the division can be absorbed into the parameters after training.

Abstract: A method and system determines in real time the bearing capacity of a foundation tamped by a high-speed hydraulic tamper. Four wireless acceleration sensors are arranged uniformly along tamping plate edges, sensor position tamping points are determined; the soil is tamped, the plate peak acceleration slows, tends to, and reaches stabilization in a range, a relationship curve between the tamping number and plate peak acceleration is determined; different loads are applied to the foundation to obtain corresponding settlements, coordinate axes are established, points are drawn according to each test data group and sequentially connected with a smooth curve to obtain a settlement-load curve, and the curve is fitted; a tamping number and foundation bearing capacity relationship is obtained; the two relationship curves are combined to obtain a relationship curve, and the foundation bearing capacity magnitude at a certain moment during the tamping operation is determined by using the acceleration index.

Abstract: Systems and methods herein describe an image compression system. The image compression system generates a first generative adversarial network (GAN), identifies a threshold, based on the threshold, generates a second GAN by pruning channels of the first GAN, trains the second GAN using similarity-based knowledge distillation from the first GAN, and stores the trained second GAN.

Abstract: A method and system determines in real time the bearing capacity of a foundation tamped by a high-speed hydraulic tamper. Four wireless acceleration sensors are arranged uniformly along tamping plate edges, sensor position tamping points are determined; the soil is tamped, the plate peak acceleration slows, tends to, and reaches stabilization in a range, a relationship curve between the tamping number and plate peak acceleration is determined; different loads are applied to the foundation to obtain corresponding settlements, coordinate axes are established, points are drawn according to each test data group and sequentially connected with a smooth curve to obtain a settlement-load curve, and the curve is fitted; a tamping number and foundation bearing capacity relationship is obtained; the two relationship curves are combined to obtain a relationship curve, and the foundation bearing capacity magnitude at a certain moment during the tamping operation is determined by using the acceleration index.

Abstract: Disclosed is a method for preparing an active Nitraria tangutorum fruit extract comprising a solvent extraction and purificaiton process and/or a supercritical extraction process. Further disclosed are a pharmaceutical composition, a food additive and a health-care preparation each comprising the active Nitraria tangutorum fruit extract obtained by the method for preparing them, and use of the extract in the manufacture for a medicament for the treatment and/or prevention of a condition.

Abstract: The present invention relates to a class of phenylpyrimidone compounds, the pharmaceutical composition, the preparation method and the use thereof. More specifically, the present invention relates to a type of phenylpyrimidone compounds of the following formula I, the pharmaceutically acceptable salts or solvates thereof and to the pharmaceutical composition as well as the preparation method of the compounds. The compounds of formula I according to the present invention can effectively inhibit type V phosphodiesterase (PDE5), and thus can be used for the treatment of various vascular disorders, such as male erectile dysfunction, pulmonary hypertension and the like.

Abstract: A reflector molding apparatus includes two pressure rollers with pressure surfaces of the rollers comprising convex teeth and concave teeth in mesh. The reflection film is put on the aluminum sheet as the unmolded material. The unmolded material goes to the pressure surfaces in mesh and is molded to be a reflector. The surface of the resulting reflector is corrugated shaped because of the pressure of the rollers. The shape of the reflector changes with the pressure surface of the rollers.

Abstract: The present invention relates to a class of phenylpyrimidone compounds, the pharmaceutical composition, the preparation method and the use thereof. More specifically, the present invention relates to a type of phenylpyrimidone compounds of the following formula I, the pharmaceutically acceptable salts or solvates thereof and to the pharmaceutical composition as well as the preparation method of the compounds. The compounds of formula I according to the present invention can effectively inhibit type V phosphodiesterase (PDE5), and thus can be used for the treatment of various vascular disorders, such as male erectile dysfunction, pulmonary hypertension and the like.

Abstract: A video apparatus has a digital encoder which receives a first analogue video signal with ancillary information in a given time window and generates on an output a digital stream based on the first analogue video signal. A digital decoder is connectable to this output to generate a second analogue video signal. A switch selects between the first analogue video signal and the second analogue video signal as an output depending on a control signal based on the occurrence of said time window.