Abstract: Provided is a system (100) for simultaneously producing a first and a second epitaxial monocrystalline layer on a respective first and second substrate, comprising a first inner container (3) defining a first cavity for accommodating a first source material and the first substrate, a second inner container (4) defining a second cavity for accommodating a second source material and the second substrate, an insulation container (6) arranged to accommodate the first and second inner containers (3, 4) therein, an outer container (7) arranged to accommodate the insulation container (6) and the first and second inner containers (3, 4) therein and heating means (8) arranged outside the outer container (7) and configured to heat the first and second cavities simultaneously.

Abstract: Aspects of the invention include systems and methods configured to provide simplified and efficient artificial intelligence (AI) model deployment. A non-limiting example computer-implemented method includes receiving an AI model deployment input having pre-process code, inference model code, and post-process code. The pre-process code is converted to a pre-process graph. The inference model and the post-process model are similarly converted to an inference graph and a post-process graph, respectively. A pipeline path is generated by connecting nodes in the pre-process graph, the inference graph, and the post-process graph. The pipeline path is deployed as a service for inference.

Abstract: A computer-implemented method for compressing a machine learning model includes converting an input machine learning model into a standard machine learning model. The method further includes converting the standard machine learning model into a plurality of pruned machine learning models, each of the pruned machine learning models converted using a corresponding pruning ratio from a pruning ratio candidate list. The method further includes determining, for each of the pruned machine learning models, a size-to-error ratio. The method further includes selecting, based on the size-to-error ratio of the pruned machine learning models, a first pruning ratio from the pruning ratio candidate list. The method further includes generating a compressed machine learning model by compressing the input machine learning model using the first pruning ratio that is selected. The method further includes deploying the compressed machine learning model for production.

Abstract: A system (100) for producing an epitaxial monocrystalline layer on a substrate (20) comprising: an inner container (30) defining a cavity (5) for accommodating a source material (10) and the substrate (20); an insulation container (50) arranged to accommodate the inner container (30) therein; an outer container (60) arranged to accommodate the insulation container (50) and the inner container (30) therein; and heating means (70) arranged outside the outer container (60) and configured to heat the cavity (5), wherein the inner container (30) comprises a support structure for supporting a solid monolithic source material (10) at a predetermined distance above the substrate (20) in the cavity (5) such that a growth surface of the substrate (20) is entirely exposed to the source material (10). A corresponding method is also disclosed.

Abstract: A computer-implemented method for training a machine learning model includes receiving a first training job at a processing device of a computer system having a plurality of graphics processing unit (GPU) resources, the first training job being part of a set of training jobs, and determining an amount of available memory in each GPU resource of the plurality of GPU resources. The method also includes loading the training job into one or more GPU resources with at least one second training job. The loading includes determining a cost model indicating an efficiency cost of each of a plurality of packing patterns, and packing the first training job and the second training job into the one or more GPU resources according to a packing pattern associated with a lowest efficiency cost.

Abstract: A roly-poly toy with adjustable center of gravity, comprising a first and second base part arranged at an interval, a first winding mandrel mounted on the first base part, a second winding mandrel mounted on the second base part, a reel belt, and a drive assembly. The reel belt is annular and has a first end wound around the first winding mandrel and a second end wound around the second winding mandrel. The drive assembly can switch between a first driving state and a second driving state. When the drive assembly is in the first driving state, the first winding mandrel rotates to roll up the reel belt, and the second winding mandrel rotates to release the reel belt. When it is in the second driving state, the second winding mandrel rotates to roll up the reel belt, and the first winding mandrel rotates to release the reel belt.

Abstract: A communication method for an SD-WAN is provided. The method may include: A first node obtains SD-WAN overlay topology information; obtains a first SD-WAN overlay locator route advertised by a second node, where the first SD-WAN overlay locator route includes a first locator configured by the second node at an overlay layer; and obtains, based on the first SD-WAN overlay locator route and the SD-WAN overlay topology information, a software-defined wide area network overlay segment routing SD-WAN overlay SR BE path arriving at the second node.

Abstract: Aspects of the invention include techniques for providing a large language model with elastic resources. A non-limiting example method includes training a large language model within a parallelized training environment that includes a set of training resources. The large language model is split into a plurality of pieces and each training resource trains over one piece. Baseline metrics are collected for the parallelized training environment. The method includes determining whether a current number of available training resources is an integer multiple of a minimum threshold of training resources and initializing one or more try-runs to evaluate vertical scaling and/or horizontal scaling responsive to the determination. A try-run having a highest improvement metric is identified and, responsive to the highest improvement metric being greater than a predetermined threshold, a training pattern is updated using one or both of a vertical scaling and a horizontal scaling of the set of training resources.

Abstract: A method of forming a semiconductor structure includes annealing a surface of a substrate in an ambient of hydrogen to smooth the surface, pre-cleaning the surface of the substrate, depositing a high-? dielectric layer on the pre-cleaned surface of the substrate, performing a re-oxidation process to thermally oxidize the surface of the substrate; performing a plasma nitridation process to insert nitrogen atoms in the deposited high-? dielectric layer, and performing a post-nitridation anneal process to passivate chemical bonds in the plasma nitridated high-? dielectric layer.

Abstract: An air volume control valve capable of accurately measuring an air volume is provided. The air volume control valve includes a round-shaped valve body, and a valve blade and an orifice plate provided in the round-shaped valve body, where the orifice plate is located upstream of the valve blade; an upstream pressure tapping port and a downstream pressure tapping port are respectively provided upstream and downstream of the orifice plate; a distance from both the upstream pressure tapping port and the downstream pressure tapping port to the orifice plate is 0.5 times a valve diameter; the valve blade is located downstream of the downstream pressure tapping port; an inlet side of the round-shaped valve body is provided with a convergent duct fitting; and the convergent duct fitting is located at a distance of once the valve diameter upstream of the orifice plate and is tapered along an air flow direction.

Abstract: An air volume control valve capable of accurately measuring an air volume is provided. The air volume control valve includes a round-shaped valve body, and a valve blade and an orifice plate provided in the round-shaped valve body, where the orifice plate is located upstream of the valve blade; an upstream pressure tapping port and a downstream pressure tapping port are respectively provided upstream and downstream of the orifice plate; a distance from both the upstream pressure tapping port and the downstream pressure tapping port to the orifice plate is 0.5 times a valve diameter; the valve blade is located downstream of the downstream pressure tapping port; an inlet side of the round-shaped valve body is provided with a convergent duct fitting; and the convergent duct fitting is located at a distance of once the valve diameter upstream of the orifice plate and is tapered along an air flow direction.

Abstract: A method, system, and computer program product for hyper-parameter determination. In a method, a network architecture of a learning model may be determined, and the learning model may be configured for performing a computing task based on machine learning. A metric value record associated with a group of hyper-parameters may be obtained during hyper-parameter determination for the learning model. An estimation of a metric value may be obtained based on the network architecture, and the metric value record and an association relationship representing an association between network architectures and metric values for the network architectures. The group of hyper-parameters may be selected in response to the estimation of the metric value meeting a predefined criterion. With these embodiments, a group of hyper-parameters may be selected, and further the learning model may be trained based on the selected group of hyper-parameters.

Abstract: A monocrystal furnace with lateral feeding includes: a furnace body, a feeding apparatus, and a thermal insulation layer, a sealing member, a crucible and a lifting apparatus that are provided within the furnace body. The thermal insulation layer is located between the crucible and the furnace body, a first through-hole is provided on a side portion of the furnace body, and a second through-hole is provided in a position of the thermal insulation layer opposite to the first through-hole; the lifting apparatus is connected to the sealing member and is used to drive the sealing member to switch between a rising state and a falling state; in the rising state, the sealing member rises, and the feeding apparatus is capable of passing through the first through-hole and the second through-hole to be opposite to an opening of the crucible.

Abstract: Methods of forming and processing semiconductor devices are described. Certain embodiments related to electronic devices which comprise a dipole region having an interlayer dielectric, a high-? dielectric material, and a dipole layer. The dipole layer comprises one or more of titanium aluminum nitride (TiAlN), titanium tantalum nitride (TiTaN), titanium oxide (TiO), tantalum oxide (TaO), and titanium aluminum carbide (TiAlC).

Abstract: Aspects of the invention include systems and methods to obtain meta features of a dataset for training in a deep learning application. A method includes selecting an initial search space that defines a type of deep learning architecture representation that specifies hyperparameters for two or more neural network architectures. The method also includes applying a search strategy to the initial search space. One of the two or more neural network architectures are selected based on a result of an evaluation according to the search strategy. A new search space is generated with new hyperparameters using an evolutionary algorithm and a mutation type that defines one or more changes in the hyperparameters specified by the initial search space, and, based on the mutation type, the new hyperparameters are applied to the one of the two or more neural networks or the search strategy is applied to the new search space.

Abstract: A system (100) for producing an epitaxial monocrystalline layer on a substrate (20) comprising: an inner container (30) defining a cavity (5) for accommodating a source material (10) and the substrate (20); an insulation container (50) arranged to accommodate the inner container (30) therein; an outer container (60) arranged to accommodate the insulation container (50) and the inner container (30) therein; and heating means (70) arranged outside the outer container (60) and configured to heat the cavity (5), wherein the inner container (30) comprises a support structure for supporting a solid monolithic source material (10) at a predetermined distance above the substrate (20) in the cavity (5) such that a growth surface of the substrate (20) is entirely exposed to the source material (10). A corresponding method is also disclosed.

Abstract: Edge device video processing with encryption of confidential image content is provided. The process includes obtaining, by the edge device, a video stream, and partitioning image content of the video stream into a confidential part and a non-confidential part. Further, the process includes encrypting, by the edge device, the confidential part of the image content to obtain encrypted image content from the confidential part and non-encrypted image content from the non-confidential part of the image content. In addition, the process includes encoding, by the edge device, the encrypted image content and the non-encrypted image content into an encoded video stream, and transmitting the encoded video stream to one or more processing servers.

Abstract: A roly-poly toy with adjustable center of gravity, comprising a first and second base part arranged at an interval, a first winding mandrel mounted on the first base part, a second winding mandrel mounted on the second base part, a reel belt, and a drive assembly. The reel belt is annular and has a first end wound around the first winding mandrel and a second end wound around the second winding mandrel. The drive assembly can switch between a first driving state and a second driving state. When the drive assembly is in the first driving state, the first winding mandrel rotates to roll up the reel belt, and the second winding mandrel rotates to release the reel belt. When it is in the second driving state, the second winding mandrel rotates to roll up the reel belt, and the first winding mandrel rotates to release the reel belt.

Abstract: A computer-implemented method for training a machine learning model includes receiving a first training job at a processing device of a computer system having a plurality of graphics processing unit (GPU) resources, the first training job being part of a set of training jobs, and determining an amount of available memory in each GPU resource of the plurality of GPU resources. The method also includes loading the training job into one or more GPU resources with at least one second training job. The loading includes determining a cost model indicating an efficiency cost of each of a plurality of packing patterns, and packing the first training job and the second training job into the one or more GPU resources according to a packing pattern associated with a lowest efficiency cost.

Abstract: Provided is a system (100) for simultaneously producing a first and a second epitaxial monocrystalline layer on a respective first and second substrate, comprising a first inner container (3) defining a first cavity for accommodating a first source material and the first substrate, a second inner container (4) defining a second cavity for accommodating a second source material and the second substrate, an insulation container (6) arranged to accommodate the first and second inner containers (3, 4) therein, an outer container (7) arranged to accommodate the insulation container (6) and the first and second inner containers (3, 4) therein and heating means (8) arranged outside the outer container (7) and configured to heat the first and second cavities simultaneously.