Abstract: Machine learning is utilized to analyze respective execution times of a plurality of tasks in a job performed in a distributed computing system to determine that a subset of the plurality of tasks are straggler tasks in the job, where the distributed computing system includes a plurality of computing devices. A supervised machine-learning algorithm is performed using a set of inputs including performance attributes of the plurality of tasks, where the supervised machine learning algorithm uses labels generated from determination of the set of straggler tasks, the performance attributes include respective attributes of the plurality of tasks observed during performance of the job, and applying the supervised learning algorithm results in identification of a set of rules defining conditions, based on the performance attributes of the plurality of tasks, indicative of which tasks will be straggler tasks in a job. Rule data is generated to describe the set of rules.

Abstract: A semiconductor structure and a method of fabricating therefor are disclosed. A second contact pad (500) is arranged lateral to a first contact pad (420) in an interconnect structure (400). As a result, during fabrication of the interconnect structure (400), the first contact pad (420) will not be present alone in a large bland area, due to the presence of the second contact pad (500). Thus, a pattern feature for the first contact pad (420) will not be over-resolved, increasing formation accuracy of the first contact pad (420) and thus guaranteeing good electrical transmission performance of the resulting interconnect structure (400).

Abstract: A liquid crystal on silicon panel including a plurality of driving circuits is provided. Each of the driving circuits includes a first driving route and a second driving route. The first driving route and the second driving route are configured to drive corresponding subpixel electrodes and respectively determine display content of a first wobulation image and a second wobulation image. In addition, a projection apparatus including the liquid crystal on silicon panel is also provided.

Abstract: A contact structure, contact pad layout and structure, mask combination and manufacturing method thereof is provided in the present invention. Through the connection of tops of at least two contact plugs in the boundary of core region, an integrated contact with larger cross-sectional area is formed in the boundary of core region. Accordingly, the process of forming electronic components on the contact structure in the boundary of core region may be provided with sufficient process window to increase the size of electronic components in the boundary, lower contact resistance, and the electronic component with increased size in the boundary buffer the density difference of circuit patterns between the core region and the peripheral region, thereby improving optical proximity effect and ensuring the uniformity of electronic components on the contact plugs inside the boundary of core region, and avoiding the collapse of electronic components on the contact plug in the boundary.

Abstract: Combinations of anti-cancer antibodies and inhibitory antibodies to CD223 overcome immune suppression in cancer patients. The inhibitory antibodies may be generated in an animal by injection of fragments of CD223. Antibodies may be monoclonal antibodies or single chain antibodies or humanized antibodies.

Abstract: Combinations of anti-cancer antibodies and inhibitory antibodies to CD223 overcome immune suppression in cancer patients. The inhibitory antibodies may be generated in an animal by injection of fragments of CD223. Antibodies may be monoclonal antibodies or single chain antibodies or humanized antibodies.

Abstract: Machine learning is utilized to analyze respective execution times of a plurality of tasks in a job performed in a distributed computing system to determine that a subset of the plurality of tasks are straggler tasks in the job, where the distributed computing system includes a plurality of computing devices. A supervised machine-learning algorithm is performed using a set of inputs including performance attributes of the plurality of tasks, where the supervised machine learning algorithm uses labels generated from determination of the set of straggler tasks, the performance attributes include respective attributes of the plurality of tasks observed during performance of the job, and applying the supervised learning algorithm results in identification of a set of rules defining conditions, based on the performance attributes of the plurality of tasks, indicative of which tasks will be straggler tasks in a job. Rule data is generated to describe the set of rules.

Abstract: Anti-CD223 antibodies overcome immune suppression in cancer patients. The anti-CD223 antibodies may be generated in an animal by injection of fragments of CD223. Antibodies may be monoclonal antibodies or single chain antibodies or humanized antibodies.

Abstract: An integrated circuit (IC) device includes a substrate having a fin-type active region extending in a first direction, a gate structure intersecting the fin-type active region on the substrate, the gate structure extending in a second direction perpendicular to the first direction and parallel to a top surface of the substrate, source and drain regions on both sides of the gate structure, and a first contact structure electrically connected to one of the source and drain regions, the first contact structure including a first contact plug including a first material and a first wetting layer surrounding the first contact plug, the first wetting layer including a second material having a lattice constant that differs from a lattice constant of the first material by about 10% or less.

Abstract: A light-emitting diode (LED) projection device for a camera has a substrate and at least one LED package mounted on the substrate. Each LED package includes a reflection surface, at least one LED chip disposed in the reflection surface, and a packaging lens encapsulating the at least one LED chip. By choosing a suitable amount of the LED chips that have suitable sizes and displacements and are encapsulated by suitable types of the packaging lenses, the light beams emitted from the LED chips can form a substantially rectangular light pattern that corresponds in shape to a field of view of the lens assembly. The brightness of the light pattern having better uniformity is formed and the dark area around the light pattern is reduced. The desired light pattern can be formed through simple manufacturing process and structure.

Abstract: A bicycle sprocket comprises a sprocket body, a sprocket ring, a plurality of limiting members, and a plurality of reinforcing members. The sprocket ring has a plurality of ring pieces, each of which includes a plurality of sprocket teeth. Through the movement of the sprocket ring relative to the sprocket body, the axially movable scope of the sprocket ring limited by the limiting members, and the binding strength between the adjacent ring pieces stabilized by the reinforcing members, the sprocket ring can shift relative to the sprocket body in order to reduce the friction and noise generated by the sprocket teeth of the ring pieces and the chain when the bicycle sprocket drives the chain to move.

Abstract: A bandgap reference circuit and method of using the same are provided. The bandgap reference circuit includes a startup component; an output component; and a bandgap core component coupled there-between. The bandgap core component includes a reference point having a voltage associated with an output signal of the output component. A controller is configured for controlling the bandgap core component and the output component to switch between a low power consumption mode and a normal operation mode based on the voltage at the reference point. When the bandgap core component and the output component operate in the normal operation mode, the bandgap reference circuit outputs a stable voltage and has a first power consumption. When the bandgap core component and the output component operate in the low power consumption mode, the bandgap reference circuit has a second power consumption less than the first power consumption.

Abstract: The present invention is related to an issue-manage-style internet public opinion information evaluation management system and method thereof. The system mainly comprises 5 modules of 1) Issue establish/setup module for establishing new issue and the keywords thereof; 2) public opinion information collection module for retrieving and analyzing data retrieved by dredge technology, program, and community web-site open API; 3) public opinion information reputation analysis module for calculating each public opinion information evaluation score by text reputation analysis and community interactive fuzzy analysis; 4) issue trend analysis module for calculating issue trend score by disclosed method based on daily, weekly, or monthly public opinion information reputation evaluation score; 5) issue related public opinion information exchange module for presenting issue related public opinion information on each management interface or message alert of each system via internet exchange standards.

Abstract: Machine learning is utilized to analyze respective execution times of a plurality of tasks in a job performed in a distributed computing system to determine that a subset of the plurality of tasks are straggler tasks in the job, where the distributed computing system includes a plurality of computing devices. A supervised machine-learning algorithm is performed using a set of inputs including performance attributes of the plurality of tasks, where the supervised machine learning algorithm uses labels generated from determination of the set of straggler tasks, the performance attributes include respective attributes of the plurality of tasks observed during performance of the job, and applying the supervised learning algorithm results in identification of a set of rules defining conditions, based on the performance attributes of the plurality of tasks, indicative of which tasks will be straggler tasks in a job. Rule data is generated to describe the set of rules.

Abstract: A Static Random Access Memory (SRAM) array power supply circuit is presented. The circuit comprises an SRAM test unit having a substantially same structure as a basic SRAM unit in the SRAM array; a switch device connected to a power source, the SRAM test unit, and the SRAM array; and a switch control circuit connected to the SRAM test unit and the switch device. When a test voltage in the SRAM test unit is lower than a threshold voltage, the switch device is closed so that the power source begins to charge the SRAM array and the SRAM test unit. The SRAM test unit provides an early warning for the SRAM array, allowing the latter to be charged upon fulfillment of a condition (e.g., charge is low). Compared to conventional circuits, this circuit provides an output voltage that is more stable and less susceptible to the changes in external conditions such as temperature or pressure.

Abstract: A light-emitting diode (LED) projection device for a camera has a substrate and at least one LED package mounted on the substrate. Each LED package includes a reflection surface, at least one LED chip disposed in the reflection surface, and a packaging lens encapsulating the at least one LED chip. By choosing a suitable amount of the LED chips that have suitable sizes and displacements and are encapsulated by suitable types of the packaging lenses, the light beams emitted from the LED chips can form a substantially rectangular light pattern that corresponds in shape to a field of view of the lens assembly. The brightness of the light pattern having better uniformity is formed and the dark area around the light pattern is reduced. The desired light pattern can be formed through simple manufacturing process and structure.

Abstract: A chainring installation structure includes a crank, a spindle, and a swing assembly. The spindle includes a first rotational axis, and one end of the spindle is connected to the crank. The crank drives the spindle to cause the spindle to rotate about the first rotational axis. The swing assembly is disposed at one end of the spindle adjacent to the crank without contacting the crank. The swing assembly can shift axially relative to the spindle. The swing assembly includes a second rotational axis that crosses the first rotational axis, in which the swing assembly can rotate about the second rotational axis.

Abstract: A flexible high-power control device is disclosed for controlling at least one high-power electrical generating/consuming device which has a plurality of electrical terminals and a housing with a non-planar surface and is thermal-conductively connected to an active cooling unit. The flexible high-power control device comprises a flexible thermal-conductive interfacial insulation substrate, at least one circuit layer, and at least one electronic element. The interfacial insulation substrate is attached to the non-planar surface of the electrical generating/consuming device. The circuit layer, which is disposed on the interfacial insulation substrate, includes a plurality of electrical terminals which are electrically connected to the electrical terminals of the electrical generating/consuming device. The electronic element is electrically mounted on the circuit layer to form a control circuit for controlling the electrical generating/consuming device.

Abstract: Method and device for detecting the process corner of a transistor are provided. The process corner detection method includes providing a ring oscillator. The ring oscillator includes an odd number of oscillation units connected in series and an output port of one of the oscillation units serves as the output port of the ring oscillator to output an oscillation signal. Each oscillation unit is constructed based on a PMOS transistor and an NMOS transistor. The process corner detection method further includes measuring the period of the oscillation signal and the maintaining time of the oscillation signal at a high level and a low level in each cycle; and determining the process corner of the PMOS transistor and the NMOS transistor in the oscillation unit based on the period of the oscillation signal and the maintaining time of the oscillation signal at a high level and a low level in each cycle.

Abstract: A method operates a bandgap voltage reference circuit that includes a bias circuit for receiving a feedback signal and outputting a bias signal, an amplifier for receiving the bias signal and outputting a first reference signal as the feedback signal, an output circuit for receiving the first reference signal and outputting a second reference signal, and an output switch for outputting the second reference signal as an output signal. The method includes, after powering up the bandgap voltage reference circuit, determining whether the output signal is stable, when the output signal is stable, turning off the output switch; turning off the bias circuit; and turning off the output circuit. The sequential turning off the output switch, the output circuit, and the bias circuit puts the bandgap voltage reference circuit into a sleep mode to save power.