Abstract: The present disclosure relates to a positioning method, a positioning server, and a positioning system. The positioning method includes: receiving a wireless positioning coordinate of a moving object; determining a visual sub-map to be matched in a visual map database based on the wireless positioning coordinate; and obtaining a visual positioning coordinates corresponding to a current image of the moving object captured at time of wireless positioning as a positioning result, based on the determined visual sub-map to be matched.

Abstract: A full-stack neural network obfuscation framework obfuscates a neural network architecture while preserving its functionality with very limited performance overhead. The framework includes obfuscating parameters or “knobs”, including layer branching, layer widening, selective fusion and schedule pruning, that increase the number of operators, reduce/increase the latency, and number of cache and DRAM accesses. In addition, a genetic algorithm-based approach is adopted to orchestrate the combination of obfuscating knobs to achieve the best obfuscating effect on the layer sequence and dimension parameters so that the architecture information cannot be successfully extracted.

Abstract: Provided are compounds represented by general formula (I), which can be used for treating ATR kinase-mediated diseases, such as proliferative diseases, such as cancer. Further provided are a pharmaceutical composition of the compound represented by general formula (I), a use thereof in treating ATR kinase-mediated diseases, and a preparation thereof.

Abstract: A robust and accurate binary neural network, referred to as RA-BNN, is provided to simultaneously defend against adversarial noise injection and improve accuracy. Recently developed adversarial weight attack, a.k.a. bit-flip attack (BFA), has shown enormous success in compromising deep neural network (DNN) performance with an extremely small amount of model parameter perturbation. To defend against this threat, embodiments of RA-BNN adopt a complete binary neural network (BNN) to significantly improve DNN model robustness (defined as the number of bit-flips required to degrade the accuracy to as low as a random guess). To improve clean inference accuracy, a novel and efficient two-stage network growing method is proposed and referred to as early growth. Early growth selectively grows the channel size of each BNN layer based on channel-wise binary masks training with Gumbel-Sigmoid function.

Abstract: Disclosed herein are system, method, and computer program product embodiments for allocating resources based on predictions of workload probability parameters. The method can include collecting a first set of historical workload data generated by operating a first set of one or more applications at a first number of past time instances; predicting probability parameters of a second set of future workload data for operating a second set of one or more applications at a second number of future time instances; and determining future resources allocated to operating the second set of one or more applications for the second number of future time instances, based on allocated current resources, a lower bound of resources to satisfy a quality of service (QoS) for operating the second set of one or more applications, an upper bound of resources to satisfy the QoS, and the predicted probability parameters.

Abstract: Embodiments relate to diagnosis and recovery of cloud based systems. From an incident ticket, a dynamic diagnostics graph is generated visualizing a hierarchy (ancestor, child) of diagnostic jobs investigating the functioning cloud system. By indicating and checking job statuses, child jobs dependent on a skipped or failed job can be skipped according to a dynamic pruning technique—thereby trimming an entire branch. And, by running separate groups of diagnostic jobs in parallel across different nodes, the diagnostic process can be finished rapidly and efficiently. A diagnostic report includes the dynamic diagnostics graph. For system recovery, the dynamic diagnostic graph is analyzed to automatically provide one or more appropriate Recommended Actions (RAs) resolving cloud system problem(s) revealed by diagnostic efforts. Those appropriate RAs may be provided by performing machine learning (e.g., referencing a neural network) with a model trained from historical cloud diagnostic and recovery activity.

Abstract: The present invention relates to a rack steel plate with a thickness up to 177.8 mm by a continuous casting slab, the constituents and mass percentages including C0.11˜0.15%, Si0.15˜0.35%, Mn0.95˜1.25%, P?0.010%, S?0.002%, Cr0.45˜0.75%, Mo0.4˜0.6%, Ni1.3˜2.6%, Cu0.2˜0.4%, Al0.06˜0.09%, V0.03˜0.06%, Nb?0.04%, N?0.006%, B0.001˜0.002%, the balance is Fe and unavoidable impurity elements. The manufacture method includes, in sequence, KR molten steel pretreatment, converter smelting, LF refining, RH refining, continuous casting through a straight-arc continuous casting machine, shielding the continuous casting slab a cover and slowly cooling, cleaning the continuous casting slab, heating, high-pressure water descaling, control rolling, straightening, slowly cooling, quenching and tempering treatment.

Abstract: An ultra-heavy steel plate with low cracking sensitivity and low yield ratio, the mass percentages of chemical components of the steel plate are C 0.05-0.09; Si 0.2-0.4; Mn 1.3-1.6; Al 0.02-0.04; Nb 0.03-0.08; V 0.03-0.08; Cr 0.1-0.5; Ni 0.1-0.5; Mo 0.1-0.3; Cu 0.2-0.5; Ti 0.01-0.02; P?0.015; S?0.003; N?0.007, the balance being Fe and inevitable impurities; the carbon equivalent is ?0.43, the cold cracking sensitivity coefficient Pcm is ?0.20. A low cracking sensitivity and low yield ratio steel plate with a thickness of 40-70 mm is manufactured by the process steps of KR molten iron pretreatment-converter smelting-LF refining-RH vacuum degassing-continuous casting-lid-covering slow cooling for the continuous casting slabs-casting slabs heating-controlled rolling-controlled cooling-hot straightening-air cooling and so on.

Abstract: An Ethernet ring protection switching (ERPS) method and a node, where the method includes clearing, by a first node, media access control (MAC) address forwarding tables of a first ERPS port and a second ERPS port, and simultaneously changing multiple ports including the first ERPS port and the second ERPS port to a broadcasting state when the first ERPS port becomes faulty, where the second ERPS port is another ERPS port except the first ERPS port, sending, in a broadcasting manner using a port in the multiple ports, the MAC packet to a target node when a MAC packet is received, and changing the multiple ports to a forwarding state, learning a MAC address according to a broadcasting path of the MAC packet to obtain a target MAC address forwarding table of the second ERPS port. Hence, ERPS efficiency may be improved.

Abstract: The present invention relates to a rack steel plate with a thickness up to 177.8 mm by a continuous casting slab, the constituents and mass percentages including C0.11˜0.15%, Si0.15˜0.35%, Mn0.95˜1.25%, P?0.010%, S?0.002%, Cr0.45˜0.75%, Mo0.4˜0.6%, Ni1.3˜2.6%, Cu0.2˜0.4%, Al0.06˜0.09%, V0.03˜0.06%, Nb?0.04%, N?0.006%, B0.001˜01002%, the balance is Fe and unavoidable impurity elements. The manufacture method includes, in sequence, KR molten steel pretreatment, converter smelting, LF refining, RH refining, continuous casting through a straight-arc continuous casting machine, shielding the continuous casting slab a cover and slowly cooling, cleaning the continuous casting slab, heating, high-pressure water descaling, control rolling, straightening, slowly cooling, quenching and tempering treatment.

Abstract: A method of arranging a capacitor array of a successive approximation register analog-to-digital converter in a successive approximation process, the method including: splitting a binary capacitor array into unit capacitors, then sorting, grouping, and rotating the original binary capacitive array involved in successive approximation conversion.

Abstract: A method of arranging a capacitor array of a successive approximation register analog-to-digital converter in a successive approximation process, the method including: splitting a binary capacitor array into unit capacitors, then sorting, grouping, and rotating the original binary capacitive array involved in successive approximation conversion.

Abstract: An anti-hydrogen-induced-cracking pressure vessel steel plate containing, chemical composition of the steel plate according to the weight percentage: 0.16-0.20% of C, 0.15-0.40% of Si, 1.05-1.20% of Mn, less than or equal to 0.008% of P, less than or equal to 0.002% of S, less than or equal to 0.01% of Nb, less than or equal to 0.01% of V, less than or equal to 0.01% of Ti, less than or equal to 0.0005% of B, and the balance being Fe and inevitable impurity elements; carbon equivalent Ceq is less than or equal to 0.42%, and the computational formula for the carbon equivalent is Ceq=C+Mn/6+(Cr+Mo+V)/5+(Ni+Cu)/15.

Abstract: The present invention provides an ultrafine-grained profile of twin-crystal wrought magnesium alloys, preparation process and use of the same. In the process, raw materials of magnesium alloys are firstly smelted and cast, and are subjected to solution treatment at 300˜500° C.; then a preform is pre-deformed, so that a great amount of twin crystal microstructure forms in the magnesium alloys and the grain size of not larger than 100 ?m is achieved; subsequently continuous ECAP process is performed at 200˜350° C., and the die is replaced in according to requirement so as to obtain the desired profile. The ultrafine-grained profile of magnesium alloys prepared in the invention has the grain sizes of from 100 to 450 nm, the tensile strength of 300˜400 MPa, and the elongation of 20˜35%. The length of the profile can be more than 10 m, meeting the needs of continuous production.

Abstract: An ultra-heavy steel plate with low cracking sensitivity and low yield ratio, the mass percentages of chemical components of the steel plate are C 0.05-0.09; Si 0.2-0.4; Mn 1.3-1.6; Al 0.02-0.04; Nb 0.03-0.08; V 0.03-0.08; Cr 0.1-0.5; Ni 0.1-0.5; Mo 0.1-0.3; Cu 0.2-0.5; Ti 0.01-0.02; P?0.015; S?0.003; N?0.007, the balance being Fe and inevitable impurities; the carbon equivalent is ?0.43, the cold cracking sensitivity coefficient Pcm is ?0.20. A low cracking sensitivity and low yield ratio steel plate with a thickness of 40-70 mm is manufactured by the process steps of KR molten iron pretreatment-converter smelting-LF refining-RH vacuum degassing-continuous casting-lid-covering slow cooling for the continuous casting slabs-casting slabs heating-controlled rolling-controlled cooling-hot straightening-air cooling and so on.

Abstract: An Ethernet ring protection switching (ERPS) method and a node, where the method includes clearing, by a first node, media access control (MAC) address forwarding tables of a first ERPS port and a second ERPS port, and simultaneously changing multiple ports including the first ERPS port and the second ERPS port to a broadcasting state when the first ERPS port becomes faulty, where the second ERPS port is another ERPS port except the first ERPS port, sending, in a broadcasting manner using a port in the multiple ports, the MAC packet to a target node when a MAC packet is received, and changing the multiple ports to a forwarding state, learning a MAC address according to a broadcasting path of the MAC packet to obtain a target MAC address forwarding table of the second ERPS port. Hence, ERPS efficiency may be improved.

Abstract: The present invention provides an ultrafine-grained profile of twin-crystal wrought magnesium alloys, preparation process and use of the same. In the process, raw materials of magnesium alloys are firstly smelted and cast, and are subjected to solution treatment at 300˜500° C.; then a preform is pre-deformed, so that a great amount of twin crystal microstructure forms in the magnesium alloys and the grain size of not larger than 100 ?m is achieved; subsequently continuous ECAP process is performed at 200˜350° C., and the die is replaced in according to requirement so as to obtain the desired profile. The ultrafine-grained profile of magnesium alloys prepared in the invention has the grain sizes of from 100 to 450 nm, the tensile strength of 300˜400 MPa, and the elongation of 20˜35%. The length of the profile can be more than 10 m, meeting the needs of continuous production.

Abstract: A method and apparatus for reducing service impairment in a LAG are provided. The method includes: dividing a service packet into multiple service sub-flows, and allocating the multiple service sub-flows to all the physical ports averagely; after a new port is added, configuring a preset quantity of service sub-flows for the new port; or, after a port is deleted, allocating the service sub-flows on the deleted port to other ports; and finding a forwarding physical port through a service packet according to the service sub-flows. An apparatus for reducing service impairment in a LAG includes: an initializing module, a new port processing module or a deleted port processing module, and a forwarding port search module. In the case of adding or deleting a port, only the affected service sub-flows are operated, thus minimizing the impact.

Abstract: A method and apparatus for reducing service impairment in a LAG are provided. The method includes: dividing a service packet into multiple service sub-flows, and allocating the multiple service sub-flows to all the physical ports averagely; after a new port is added, configuring a preset quantity of service sub-flows for the new port; or, after a port is deleted, allocating the service sub-flows on the deleted port to other ports; and finding a forwarding physical port through a service packet according to the service sub-flows. An apparatus for reducing service impairment in a LAG includes: an initializing module, a new port processing module or a deleted port processing module, and a forwarding port search module. In the case of adding or deleting a port, only the affected service sub-flows are operated, thus minimizing the impact.