Abstract: Disclosed solutions for improved machine learning (ML) employ knowledge balancing self-distillation with adaptive mutual information (AMI). Examples include: for a neural network (NN) having a plurality of modules, determining a task objective for at least a final module of the plurality of modules; for the NN, determining a balancing objective using at least an output of the final module and an output of a first intermediate module of the plurality of modules; determining an overall objective, wherein determining the overall objective comprises combining the task objective with the balancing objective; and adjusting weights of the NN to minimize the overall objective. Balancing information may combine mutual information (between an intermediate module output and the output of the final module) with self-information (for the intermediate module output) to produce AMI. Adjusting weights of the NN during training, using the AMI, results in knowledge balancing self-distillation.

Abstract: A classifier model is trained for temporal action localization of video clips. A training video clip that includes actions of interest for identification is ingested into the classifier model. Action characteristics within frames of the video clip are identified. The actions correspond to known action classes. An actionness score is determined for each of the frames based upon the action characteristics identified within each of the frames. Class activation sequence (CAS) scores are determined for sequences of the frames based upon a presence or an absence of the action characteristics identified within each of the frames. Base confidence predictions of temporal locations of actions of interest within the video clip are produced by correlating each of the actionness scores with corresponding class activation scores for each of the frames in the sequences of frames.

Abstract: A neural architecture search (NAS) with a weak predictor comprises: receiving network architecture scoring information; iteratively sampling a search space, wherein the sampling comprises: generating a set of candidate architectures within the search space; learning a first predictor; evaluating performance of the candidate architectures; and based on at least the performance of the set of candidate architectures and the network architecture scoring information, refining the search space to a smaller search space; based on at least the network architecture scoring information, thresholding the performance of candidate architectures to determine scored output candidate architectures; and reporting the scored output candidate architectures. In some examples, the candidate architectures each comprise a machine learning (ML) model, for example a neural network (NN).

Abstract: A positive electrode material has a chemical general formula of LiaNixCoyMnzO2·cA·dB, in which 1.00?a?1.20, 0.00<c?0.01, 0.00<d?0.02, 0.00?x<1.00, 0.00?y<0.2, 0.00?z<0.4, and x+y+z=1; A is a first coating material, and B is a second coating material; and the first coating material A is a carbon-coated fast-ionic conductor, and the second coating material B is a carbon escape-prevention compound. A fast-ionic conductor modified through carbon coating is used to modify a positive electrode material, such that the ionic conductivity and the electronic conductivity of the positive electrode material are synchronously improved. The surface of the carbon-coated fast-ionic conductor is coated with boric acid due to the glassy property, such that the escape of carbon in the carbon-coated fast-ionic conductor is effectively prevented.

Abstract: The present disclosure provides a large single-crystal Prussian white-based sodium-ion positive electrode material and a method for preparing same. The Prussian white-based sodium-ion positive electrode material has a chemical general formula NaxMn[Fe(CN)6], wherein 1.9?x<2.00. The method comprises: (1) mixing and dissolving Na4Fe(CN)6·10H2O and ethylenediaminetetraacetic acid manganese sodium salt to prepare a first solution A; and (2) dropwise adding a diluted acid, as a second solution B, into the first solution A, performing an aging reaction, washing, and drying to give the Prussian white-based sodium-ion positive electrode material with a low water content. By using an acidic cation dissociation method, the present disclosure synthesizes a Prussian white-based sodium-ion positive electrode material with a low water content at room temperature, and the synthesized material features large primary particles, good thermal stability, a low water content, and a high capacity.

Abstract: Example solutions for video object segmentation (VOS) use a bilateral attention transformer in motion-appearance neighboring space, and perform a process that includes: receiving a video stream comprising a plurality of video frames in a sequence; receiving a first object mask for an initial video frame of the plurality of video frames; selecting a video frame of the plurality of video frames as a current query frame, the current query frame following, in the sequence, a reference frame of a reference frame set, wherein each reference frame has a corresponding object mask; using the current query frame and a video frame in the reference frame set, determining a bilateral attention; and using the bilateral attention, generating an object mask for the current query frame.

Abstract: A single-crystal cathode material and preparing method thereof are provided. The method involves mixing and ball milling a lithium source with a nickel-cobalt-manganese precursor and then performing a first sintering treatment to obtain the first main material. The first sintering temperature is 650 to 950° C. and sintering time is 15 to 30 hours. The first main material is then mixed and ball-milled with source A and performing a second sintering treatment to prepare the single-crystal cathode material. The second sintering temperature is 650-950° C. and sintering time is 5-15 hours. A precursor is used to directly prepare a single-crystal cathode material without jet milling.

Abstract: Embodiments of the present application relate to a high-nickel compound and a preparation method therefor. According to one embodiment of the present application, the high-nickel compound has a chemical general formula of LiaNixCoyMnzMbO2·c?·d?, where 1?a?1.2, 0<b?0.01, 0<c?0.01, 0<d?0.02, 0.8?x?1, 0?y<0.12, 0?z?0.2, and x+y+z=1; M is a doping element; ? is a first coating material, and ? is a second coating material. Some other embodiments of the present application further provide a method for preparing a high-nickel compound. The high-nickel compound and the preparation method therefor provided by the embodiments of the present application can effectively solve the problems encountered in the traditional technology.

Abstract: The present disclosure provides cannabicyclol derivatives and preparation methods thereof, belonging to the field of biomedical technology. In the present disclosure, citral (3,7-Dimethyl-2,6-octadienal) and derivatives thereof, and olive alcohol (1,3-Dihydroxy-5-pentylbenzene) and derivatives thereof are used as raw materials for the efficient synthesis of the cannabicyclol derivatives in a one-pot manner under an action of ethylenediamine and a photocatalyst.

Abstract: A novel biomimetic nano-sphere M@P-WI includes PLGA nanoparticles and the membrane of breast cancer EO771 cells. The PLGA nanoparticles are encapsulated within the membrane of breast cancer EO771 cells, and IR780 and DDR2 inhibitor WRG-28 are embedded in the PLGA nanoparticles. A method for preparing and applying the novel biomimetic nano-sphere M@P-WI is also provided. The method utilizes the aforementioned novel biomimetic nano-sphere M@P-WI, to trigger the conversion of light into heat energy upon near-infrared laser irradiation. This “burns” the tumor and induces specific immune responses within the body, inhibiting tumor recurrence and metastasis. Under the action of WRG-28, it promotes apoptosis of cancer-associated fibroblasts (CAF), reduces the remodeling of tumor extracellular matrix microenvironment, enhances the distribution of nanomaterials within the tumor, and restricts the invasiveness of breast cancer cells.

Abstract: Systems, apparatuses, methods, and computer program products are disclosed for distillation of a natural language processing model. An example method includes receiving, by communications circuitry, a set of text data comprising a set of observations and predicting, by processing circuitry and using the NLP model, classifications for each observation in the text data. The example method further includes generating, by model training engine, a balanced sampled data structure based on the predicted classifications for each observation in the text data and training, by the model training engine, a surrogate model using the balanced sampled data structure. The example method further includes identifying, by an interpreter and from the surrogate model, a set of most-influential tokens in the text data.

Abstract: Disclosed is a dynamic regulation method for a lane changing decision point of a CAV dedicated lane in a diverging area of an expressway, applied to the diverging area of the expressway in which the CAV dedicated lane is provided at an inner side of a road. Vehicles traveling in the CAV dedicated lane are all CAVs. Both the CAVs and human-driven vehicles can travel in ordinary lanes. The method calculates the lane changing decision points of diversion vehicles in the CAV dedicated lane at current time using corresponding algorithms based on a relationship between the lane changing decision points of the CAV dedicated lane and traffic efficiency and a traffic safety.

Abstract: Disclosed is an example approach in which network and non-network features are used to train a predictive machine learning model that is implemented to predict financial crime and fraud. Graphical network features may be generated by applying financial entity risk vectors to a network model with representations of various types of networks. The network model may comprise transactional, non-social, and/or social networks, with edges corresponding to linkages that may be weighted according to various characteristics (such as frequency, amount, type, recency, etc.). The graphical network features may be fed to the predictive model to generate a likelihood and/or prediction with respect to a financial crime. A perceptible alert is generated on one or more computing devices if a financial crime is predicted or deemed sufficiently likely. The alert may identify a subset of the set of financial entities involved in the financial crime and present graphical representations of networks and linkages.

Abstract: A computer device for automatic feature detection comprises a processor, a communication device, and a memory configured to hold instructions executable by the processor to instantiate a dynamic convolution neural network, receive input data via the communication network, and execute the dynamic convolution neural network to automatically detect features in the input data. The dynamic convolution neural network compresses the input data from an input space having a dimensionality equal to a predetermined number of channels into an intermediate space having a dimensionality less than the number of channels. The dynamic convolution neural network dynamically fuses the channels into an intermediate representation within the intermediate space and expands the intermediate representation from the intermediate space to an expanded representation in an output space having a higher dimensionality than the dimensionality of the intermediate space.

Abstract: Example solutions for video frame action detection use a gated history and include: receiving a video stream comprising a plurality of video frames; grouping the plurality of video frames into a set of present video frames and a set of historical video frames, the set of present video frames comprising a current video frame; determining a set of attention weights for the set of historical video frames, the set of attention weights indicating how informative a video frame is for predicting action in the current video frame; weighting the set of historical video frames with the set of attention weights to produce a set of weighted historical video frames; and based on at least the set of weighted historical video frames and the set of present video frames, generating an action prediction for the current video frame.

Abstract: An efficiency engine identifies container sizes for containers of a workload and allocates the containers across server clusters and nodes based on peak resource usage requirements of the containers. Runtime feedback signals are generated from monitors within the containers indicative of a quality of service and resource usage. A decision engine can identify a bin packing action to take based upon the runtime feedback signals, and a control plane can perform the identified bin packing actions to adjust bin packing based upon the runtime feedback signals. Also, adaptive adjustment can be performed based on feedback signals and using a prediction engine.

Abstract: A classifier model is trained for temporal action localization of video clips. A training video clip that includes actions of interest for identification is ingested into the classifier model. Action characteristics within frames of the video clip are identified. The actions correspond to known action classes. An actionness score is determined for each of the frames based upon the action characteristics identified within each of the frames. Class activation sequence (CAS) scores are determined for sequences of the frames based upon a presence or an absence of the action characteristics identified within each of the frames. Base confidence predictions of temporal locations of actions of interest within the video clip are produced by correlating each of the actionness scores with corresponding class activation scores for each of the frames in the sequences of frames.

Abstract: Disclosed is a dynamic regulation method for a lane changing decision point of a CAV dedicated lane in a diverging area of an expressway, applied to the diverging area of the expressway in which the CAV dedicated lane is provided at an inner side of a road. Vehicles traveling in the CAV dedicated lane are all CAVs. Both the CAVs and human-driven vehicles can travel in ordinary lanes. The method calculates the lane changing decision points of diversion vehicles in the CAV dedicated lane at current time using corresponding algorithms based on a relationship between the lane changing decision points of the CAV dedicated lane and traffic efficiency and a traffic safety.

Abstract: Embodiments of the present disclosure provide a method for traffic flow control considering a bus stop in a connected environment, comprising: determining an influence region of the bus stop; dividing the influence region of the bus stop into influence subregions; collecting vehicle information of each of the influence subregions at time t; comparing lane changing pressures of different influence subregions of the bus stop; determining a predicted count of lane changing vehicles of each of the influence subregions based on lane changing pressure differences; and selecting optimal lane changing vehicles to complete lane changing. A vehicle density distribution in the influence region of the bus stop can be dynamically regulated through vehicle information interaction in the connected environment, thereby making the spatial distribution of the traffic flow more balanced, saving vehicle traveling time, and improving the operation efficiency of the traffic flow.

Abstract: The present invention relates to an auditing system for a built environment of an age-friendly street based on multisource big data. The auditing system includes: a data acquisition module is configured to acquire urban streetscape image data, urban road network data and urban point-of-interest data; a data classification auditing module is configured to acquire the data of the data acquisition module, classify the image data, and process the image data by using a data processing method to acquire evaluated numerical values of different types of indexes; a data summary analysis module is configured to acquire the evaluated numerical values of the data classification auditing module, calculate sub-item index numerical values of each output unit and calculate result data according to the sub-item index numerical values; a audit result output module is configured to acquire the result data of the data summary analysis module and visualize and output the result data.