Abstract: A server system may receive two or more Quality of Service (QoS) dimensions for the multi-objective optimization model, wherein the two or more QoS dimensions include at least a first QoS dimension and a second QoS dimension. The server system may maximize the multi-objective optimization model along the first QoS dimension, wherein the maximizing includes selecting one or more pipelines for the multi-objective optimization model in the software architecture that meet QoS expectations specified for the first QoS dimension and the second QoS dimension, wherein an ordering of the pipelines is dependent on which QoS dimensions were optimized and de-optimized and to what extent, wherein the multi-objective optimization model is partially de-optimized along the second QoS dimension in order to comply with the QoS expectations for the first QoS dimension, and whereby there is a tradeoff between the first QoS dimension and the second QoS dimension.

Abstract: The present disclosure relates to systems and methods for a self-adjusting corporation-wide discovery and integration feature of a machine learning system that can review a client's data store, review the labels for the various data schema, and effectively map the client's data schema to classifications used by the machine learning model. The various techniques can automatically select the features that are predictive for each individual use case (i.e., one client), effectively making a machine learning solution client-agnostic for the application developer. A weighted list of common representations of each feature for a particular machine learning solution can be generated and stored. When new data is added to the data store, a matching service can automatically detect which features should be fed into the machine-learning solution based at least in part on the weighted list. The weighted list can be updated as new data is made available to the model.

Abstract: The present disclosure relates to systems and methods for a machine-learning platform for the safe serialization of a machine-learning application. Individual library components (e.g., a pipeline, a microservice routine, a software module, and an infrastructure model) can be encrypted using one or more keys. The keys can be stored in a location different from the storage location of the machine-learning application. Prior to incorporation of the library component into a machine-learning model, one or more keys can be retrieved from the remote storage location to authenticate that the one or more encrypted library components are authentic. The process can reject any of the one or more component, when the encrypted library component fails authentication. If a component is rejected, the system can roll back to a previous, authenticated version of the library component. The authenticated library components can be compiled into machine-learning software.

Abstract: The present disclosure relates to systems and methods for using existing data ontologies for generating machine learning solutions for a high-precision search of relevant services to compose pipelines with minimal human intervention. Data ontologies can be used to create a combination of non-logic based and logic-based sematic services that can significantly outperform both kinds of selection in terms of precision. Quality of Service (QoS) and product Key Performance Indicator (KPI) constraints can be used as part of architecture selection in developing, training, validating, and improving machine learning models. For data sets without existing ontologies, one or more ontologies be generated and stored for future use.

Abstract: The present disclosure relates to systems and methods for an intelligent assistant (e.g., a chatbot) that can be used to enable a user to generate a machine learning system. Techniques can be used to automatically generate a machine learning system to assist a user. In some cases, the user may not be a software developer and may have little or no experience in either machine learning techniques or software programming. In some embodiments, a user can interact with an intelligent assistant. The interaction can be aural, textual, or through a graphical user interface. The chatbot can translate natural language inputs into a structural representation of a machine learning solution using an ontology. In this way, a user can work with artificial intelligence without being a data scientist to develop, train, refine, and compile machine learning models as stand-alone executable code.

Abstract: A server system may match a segment of code for a code integration request to metadata about similar segments of code, wherein the metadata qualifies one or more outcomes of previous integration requests. The server may determine usage rights and rules based on the metadata, wherein some of the usage rights and rules have previously have been approved by a multi-approval workflow that enforces a predetermined process to authorize use of the segment of code for code segment integrations. The server may analyze the metadata to predict an integration score based at least in part on the usage rights and rules of the segments of code. If the integration score of the segment of code exceeds a threshold, the system may automatically generate a data structure for deploying the segment of code, wherein the automatically generating the data structure is performed without the multi-approval workflow.

Abstract: The present disclosure relates to systems and methods for a machine learning platform that generates a library of components to generate machine learning models and machine learning applications. The machine learning infrastructure system allows a user (i.e., a data scientist) to generate machine learning applications without having detailed knowledge of the cloud-based network infrastructure or knowledge of how to generate code for building the model. The machine learning platform can analyze the identified data and the user provided desired prediction and performance characteristics to select one or more library components and associated API to generate a machine learning application. The machine learning can monitor and evaluate the outputs of the machine learning model to allow for feedbacks and adjustments to the model. The machine learning application can be trained, tested, and compiled for export as stand-alone executable code.

Abstract: Systems and methods for determining an order accepting mode for a target service provider are provided. A method includes receiving a request for determining an order accepting mode for the target service provider. The method also includes obtaining an order accepting evaluation parameter with respect to the target service provider in response to the request. The method also includes determining an order accepting mode for the target service providers based on the order accepting evaluation parameter. The order accepting mode includes at least one of an order designating mode, an order striving mode, or an order quick-accepting mode.

Abstract: An auxiliary sputum excretion method for a non-invasive ventilator and the invasive ventilator, solving the technical problem of how to help a patient with non-invasive ventilation to excrete sputum. In the auxiliary sputum excretion method, a sputum excretion operation is initiated in each respiratory cycle of a patient after the non-invasive ventilator determines that the patient needs sputum excretion during a normal mechanical ventilation of the non-invasive ventilator for the patient, and the sputum excretion operation is stopped in response to determining that the patient needs to stop the sputum excretion.

Abstract: A system and method for determining a team are provided. The method may include obtaining information relating to each of a plurality of participants to be grouped, the plurality of participants including a plurality of team leaders and a plurality of team members. The method may also include determining one or more teams by automatically allocating the plurality of team members to the plurality of team leaders based on the information relating to the plurality of participants, wherein each of the one or more teams includes a team leader and at least one team member.

Abstract: Systems and methods involving nanomaterial-based electrodes, such as supercapacitor and battery electrodes that can be flexible, are described.

Abstract: The disclosure discloses a visual multi-object tracking based on multi-Bernoulli filter with YOLOv3 detection, belonging to the fields of machine vision and intelligent information processing. The disclosure introduces a YOLOv3 detection technology under a multiple Bernoulli filtering framework. Objects are described by using anti-interference convolution features, and detection results and tracking results are interactively fused to realize accurate estimation of video multi-object states with unknown and time-varying number. In a tracking process, matched detection boxes are combined with object tracks and object templates to determine new objects and re-recognize occluded objects in real time. Meanwhile, under the consideration of identity information of detected objects and estimated objects, identity recognition and track tracking of the objects are realized, so that the tracking accuracy of the occluded objects can be effectively improved, and track fragments are reduced.

Abstract: The present invention relates to an azolethione flotation collector and application thereof. According to the application, an azolethione compound such as a 1,3,4-thiadiazole-2-thione compound, a 1,3,4-oxadiazole-2-thione compound, a 1,2,4-triazole-3-thione compound or a 1,2,4,5-tetrazole-3-thione compound is used as a mineral flotation collector to be applied to ores containing copper, zinc, lead, nickel, cobalt, platinum, palladium, silver or gold minerals to realize flotation recovery of valuable metal minerals. Compared with common flotation collectors in the existing technologies, the flotation collector of the present invention can effectively improve enrichment and recovery of copper, zinc, lead, nickel, cobalt, platinum, palladium, silver or gold minerals.

Abstract: A superhard polycrystalline construction comprises a body of polycrystalline superhard material comprising a structure comprising superhard material, the structure having porosity greater than 20% by volume and up to around 80% by volume. A method of forming such a superhard polycrystalline construction comprises forming a skeleton structure of a first material having a plurality of voids, at least partially filling some or all of the voids with a second material to form a pre-sinter assembly, and treating the pre-sinter assembly to sinter together grains of superhard material to form a body of polycrystalline superhard material comprising a first region of superhard grains, and an interpenetrating second region; the second region being formed of the other of the first or second material that does not comprise the superhard grains; the superhard grains forming a sintered structure having a porosity greater than 20% by volume and up to around 80% by volume.

Abstract: The present disclosure is directed to significantly improving the adiabatic shear banding susceptibility of pure refractory metals as well as overcoming the physical dimension limitations when making kinetic energy penetrators. These improvements may be achieved by arranging interlayers between plasticly deformed refractory metal material layers. Disclosed herein are methods of making material for kinetic energy penetrator applications, the methods comprising: severely plasticly deforming a refractory metal material until the grain size of the refractory metal material is within one of ultrafine grain and nanocrystalline regimes; arranging an interlayer material adjacent the refractory metal material; and diffusion bonding the interlayer material to the refractory metal material.

Abstract: The present disclosure provides a method and a device for controlling a vehicle. The vehicle is in a vehicle queue, the vehicle queue includes n vehicles {vehicle 0, vehicle 1, . . . , vehicle n?1} arranged in order along a traveling direction, the method is performed by a computing device of the vehicle i, i is from 1 to n?1. A vehicle-following subsystem model is established based on the relative speed and the relative vehicle distance error between the vehicle i and a vehicle i?1, and the acceleration of the vehicle i, a state trajectory of the vehicle-following subsystem model on a ?v-?R plane is determined, a plurality of division graphs of the vehicle i?1 in a plurality of travelling phases are combined to obtain a switching control graph for the vehicle i, and the travelling phase of the vehicle i is adjusted.

Abstract: A preparation method for a platform-shaped active region based P-I-N diode string in a reconfigurable loop antenna includes: (a) selecting an SOI substrate; (b) etching the SOI substrate to form a platform-shaped active region; (c) depositing a P-type Si material and an N-type Si material around the platform-shaped active region by an in-situ doping process to form a P region and an N region respectively; (d) depositing a polysilicon material around the platform-shaped active region; (e) forming leads on a surface of the polysilicon material and forming PADs by photolithography, to form the P-I-N diode string. Therefore, a high-performance platform-shaped active region based P-I-N diode string suitable for a solid-state plasma antenna can be provided by an in-situ doping process.

Abstract: Systems and methods for determining an order accepting mode for a target service provider are provided. A method includes receiving a request for determining an order accepting mode for the target service provider. The method also includes obtaining an order accepting evaluation parameter with respect to the target service provider in response to the request. The method also includes determining an order accepting mode for the target service providers based on the order accepting evaluation parameter. The order accepting mode includes at least one of an order designating mode, an order striving mode, or an order quick-accepting mode.

Abstract: A mobile device camera stand includes a flat, planar base comprising a platform for receiving a document, an adjustable shaft projecting from the base at a first end of the shaft, and a flat, planar cradle having a top surface and a bottom surface, the bottom surface connected to a second end of the shaft. The top surface of the cradle receives a mobile device having a camera, and the cradle includes a cutout such that the camera is placed on the cutout such that the cutout does not obstruct the camera, thereby enabling the camera to image the document.

Abstract: A termination assembly for an umbilical comprising a chamber through which pass one or more tubes extending from the umbilical, the chamber containing a filler material, wherein at least one tube is provided with an anchor for securing the at least one tube in the filler material, the anchor comprising a shape memory material. A method of making a termination assembly for an umbilical. The method comprises the steps of: (a) providing an anchor for mounting onto one of the tubes, the anchor comprising a shape memory material in a second phase and comprising an aperture which is in a second shape which the tube can pass through, (b) sliding the anchor onto the tube by passing the tube through the aperture, (c) allowing the shape memory alloy to undergo a phase change in which at least some of the second phase converts to a first phase such that the aperture returns towards a first shape which the tube cannot pass through, the anchor thereby gripping the tube, and (d) adding a filler material to the chamber.