Abstract: Systems and methods are provided for building a user model. The system includes a processor and a non-transitory storage medium accessible to the processor. The processor is configured to obtain user data from a database, where the user data include user behavior for a plurality of apps installed on one or more user terminals. The processor selects at least one rating parameters using the user data, where the at least one rating parameters indicates a rating of relevant app usage. The system builds the user model based on a rating matrix comprising the at least one rating parameters.

Abstract: Methods, systems, and computer program products for providing the status of model extraction in the presence of colluding users are provided herein. A computer-implemented method includes generating, for each of multiple users, a summary of user input to a machine learning model; comparing the generated summaries to boundaries of multiple feature classes within an input space of the machine learning model; computing correspondence metrics based at least in part on the comparisons; identifying, based at least in part on the computed metrics, one or more of the multiple users as candidates for extracting portions of the machine learning model in an adversarial manner; and generating and outputting an alert, based on the identified users, to an entity related to the machine learning model.

Abstract: Methods and apparatus for reducing a size of a neural network model, the method including: compressing data of the neural network model; identifying structure information of a vector register, wherein the structure information includes a number of registers included in the vector register; comparing a number of elements in the compressed data with a first condition, wherein the first condition is determined based on the number of registers in the vector register; and in response to the number of elements satisfying the first condition, associating the compressed data with the vector register to enable loading the compressed data to the vector register.

Abstract: Disclosed herein is a mounting mechanism such as a drift tube mounting mechanism for mounting a drift tube provided in a drift tube accelerator vessel 10 via an elongate holding element 14. The mounting mechanism comprises a biasing mechanism comprising one or more disc spring members 46 for axially biasing the elongate holding element 14 with respect to the vessel towards a mounting position. The mounting mechanism further comprises a screw 58 having a head portion 60 and a shaft portion, said shaft portion comprising a threaded portion 64 for screwing into the elongate holding element 14. The mounting mechanism further comprises a bushing 66 comprising a sleeve portion 68 receiving at least a part of the screw shaft and separating the one or more disc spring members 46 from said screw shaft portion. The screw head portion 60 is configured to receive the axial biasing force generated by the one or more disc spring members 46.

Abstract: One embodiment of the present invention sets forth a mechanism for transmitting and receiving differential signals. A transmitter combines a direct current (DC) to DC converter including a capacitor with a 2:1 multiplexer to drive a pair of differential signaling lines. The transmitter drives a pair of voltages that are symmetric about the ground power supply level. Signaling currents are returned to the ground plane to minimize the generation of noise that is a source of crosstalk between different differential signaling pairs. Noise introduced through the power supply is correlated with the switching rate of the data and may be reduced using an equalizer circuit.

Abstract: Embodiments of the present disclosure relate to techniques for controlling the temperature of light sources within physiological sensors in order to regulate the wavelengths emitted by the light sources. The sensors may include a temperature control element that is designed to provide heating and/or cooling to the light sources. The sensors also may include a temperature sensor designed to detect the temperature of the light sources. Based on the detected temperature, a controller can vary the amount of heating and/or cooling provided by the temperature control element to maintain the temperature of the light sources at a desired temperature or within a desired temperature range.

Abstract: For signal luminaires which indicate the term STOP or DANGER upon activation, a circuit containing a first operating case feed unit and a second auxiliary source is provided for the purpose of supplying a reliable power supply. In this case, the second auxiliary source is inductively coupled to the signal luminaire electric circuit; while the signal luminaires are capacitively coupled to ground. The use of switches during operation in a fall-back level is obviated in this way.

Abstract: A pre-emphasis circuitry that includes (1) a pre-emphasis voltage variation compensation (PVVC) engine having a transition detection circuit and (2) a compensation driver coupled to the PVVC engine is described. In one embodiment, the compensation driver reduces data dependent voltage variations in pre-emphasis provided by the pre-emphasis circuitry. In one embodiment, in response to a predetermined data pattern detected by the PVVC engine, the compensation driver provides an additional boost to performance critical capacitive nodes of the pre-emphasis circuitry. The additional boost causes the performance critical capacitive nodes to charge or discharge more rapidly. In one embodiment, the PVVC engine further includes a digital finite impulse response (FIR) filter coupled to the transition detection circuit.

Abstract: A constant current circuit includes a first transistor, a second transistor having the gate and the source connected to the gate and the source of the first transistor, and having the drain connected to a load, a voltage adjustment circuit section that controls the drain voltage of the first transistor, a constant current generation circuit section that supplies a constant current to the first transistor, and a detection circuit section that determines whether at least one of the first transistor and the second transistor is unable to output a current proportional to the first constant current while at least one of the first transistor and the second transistor operates in the linear region, by performing a voltage comparison between a voltage at a connecting section between the voltage adjustment circuit section and the constant current generation circuit section and a predetermined reference voltage.

Abstract: A through silicon via (TSV) repair circuit of a semiconductor apparatus is provided. The TSV repair circuit includes a first chip, at least one second chip, at least two TSVs, at least two data path circuits, and an output logic circuit. Each data path circuit includes an input driving circuit, a short-circuit detection circuit, a bias circuit, and a leakage current cancellation circuit. The input driving circuit transforms an input signal into a pending signal and transmits the pending signal to a first terminal of the corresponding TSV. The short-circuit detection circuit detects a short circuit between the corresponding TSV and a silicon substrate according to the input signal and the first terminal of the TSV and generates a short-circuit detection output signal. The leakage current cancellation circuit prevents a leakage current produced by a first level voltage from entering the silicon substrate according to the short-circuit detection output signal.

Abstract: A dielectric window for a plasma treatment device for a plasma treatment device that uses microwaves as a plasma source. The dielectric window is circular-plate-shaped and allows microwaves to propagate. The dielectric window has a recess that has an opening on the lower-surface side and that indents in the plate thickness direction of the dielectric window, and is provided to the lower surface at which plasma is generated when the dielectric window is provided to the plasma treatment device. The recess has a bottom surface extending in the direction perpendicular to the plate thickness direction, and a side surface extending in the plate thickness direction from the circumferential edge of the bottom surface toward the opening of the recess. In addition, an inclined surface extends at an incline relative to the plate thickness direction from the opening-side circumferential edge of the side surface toward the opening of the recess.

Abstract: An impedance tuning circuit includes a calibration unit and a post-processing unit. The calibration unit generates an initial pull-up code and an initial pull-down code by performing a calibration operation using an external resistor during an initial impedance tuning operation. The post-processing unit outputs the initial pull-up code and the initial pull-down code as a final pull-up code and a final pull-down code during the initial impedance tuning operation, and generates the final pull-up code and the final pull-down code by using the initial pull-up code and the initial pull-down code during a subsequent impedance tuning operation.

Abstract: Preservation of quantum entanglement in a two-qubit system is achieved by use of the disclosed systems. Three different example two-qubit systems are shown: (1) a system employing a weak measurement, (2) a system in which a generalized amplitude dampening occurs without use of a weak measurement, and (3) an extended system in which the system is prepared in a more robust state less susceptible to decoherence prior to a generalized amplitude dampening.

Abstract: An electronic logic circuit uses areal capacitive coupling devices coupled together to process a set of data inputs. Each areal capacitive coupling device can be configured such that a floating gate potential of such device can be altered to at least a first state or a second state in response to receiving an input signal from the set of data inputs, which is coupled electrically to the floating gate. A majority function logic circuit (and other similar circuits) can be interconnected this way using far fewer gates than with a conventional CMOS implementation. Selective logic gates can also be enabled or disabled by configuring them effectively as memory devices.

Abstract: A photoelectric sensing sensitivity adjusting device includes: an LED for emitting light by being supplied with energy (current or voltage); a regulator for LED driving for supplying the energy to the LED; a light receiving device for receiving the light emitted from the LED; a digital potentiometer for controlling increase or decrease of the energy to be supplied to the LED; a CPU for digitally controlling, according to a control program, the increase or decrease of the energy to be supplied to the LED; and a photodetection portion for detecting an optical signal, converting the optical signal to an electrical signal, and outputting the electrical signal.

Abstract: Circuitry comprises a high voltage rail providing a high voltage level corresponding to a higher voltage domain, an intermediate voltage source, a low voltage rail, and devices that operate in a lower voltage domain. First devices in an upper voltage region are powered between the high voltage rail and an intermediate voltage rail powered by the intermediate source. Second devices in a lower voltage region are powered between the intermediate and low rails. On power up, the intermediate source is powered before the high voltage rail. An isolating circuit connects the intermediate source to a node when the high voltage rail is powered and isolates the node from the intermediate source when the high voltage rail is not powered to impede current flow from the intermediate source to the high voltage rail.

Abstract: A voltage-driven pixel circuit, a driving method thereof and a display panel including the voltage-driven pixel circuit are disclosed. The voltage-driven pixel circuit comprises a driving transistor, a retaining transistor, a switching transistor, a compensating transistor, a storage capacitance and an OLED device. The technical solutions disclosed here compensate for the unevenness of the threshold voltage of the N-type TFT transistors and OLED efficiently.

Abstract: An example power converter includes a first winding, a second winding, a switch, a controller and an output circuit. The second winding is magnetically coupled to the first winding and the controller includes a feedback terminal and a common terminal. The controller is coupled to control the switch to regulate an output of the power converter in response to a feedback voltage received at the feedback terminal. The output circuit is coupled between the common terminal of the controller and a common reference of the power converter to provide an output voltage to a load. The feedback voltage is a positive voltage with respect to the common terminal and the output voltage is a negative voltage with respect to the common reference of the power converter.

Abstract: Circuits and methods prevent ringing of white LED drivers by adding a high voltage coil switch (CS) circuit, a high voltage edge detector circuit, and a boot strap circuit. The coil switch (CS) circuit to short the coil of a boost converter, when the current in it reaches zero during DCM, comprises a series of 2 identical HV enhancement NMOS with bulk switch system. An Edge Detector circuit senses the fast falling edge of an LX node. A Boost Strap circuit lifts the gate of coil switch circuit above the battery voltage level and turns on the coil switch. A Boost Strap (BS) system lifts the gate of CS above the VBAT level and turns on CS, wherein minimum current is required to minimize impact on efficiency.

Abstract: Disclosed herein is a remotely adjustable lighting system with security features. The lighting system includes a lamp (e.g., luminaire) which carries solid-state light sources and a sensor. The solid-state light sources emit visible light at an adjustable output intensity. The sensor is responsive to line-of-sight wireless signals, and the lamp increases and decreases the output intensity of the light sources in response to output signals of the sensor. The lamp provides security features by selectively accepting and rejecting instructions carried by the line-of-sight wireless signals to vary the output intensity of the light sources.