Abstract: A computer-implemented method for training a representation learning model to be able to determine predictive maintenance data from irregular-sampled, and variable-length timeseries data includes providing unlabeled timeseries data indicative of a state of a device under surveillance; embedding the timeseries data that generates embedded timeseries data indicative of the relative temporal distance of the entries relative to each other; performing a first training of the representation learning model by masking a predetermined number of temporally consecutive pieces of observation data; attaching to the representation learning model a fully-connected layer that normalizes a representation learning model output and feeds the normalized output to a loss model that is indicative of a specific predictive maintenance task; and performing a second training of the representation learning model based on the loss model and sparsely labelled timeseries data in order to obtain a trained representation learning model for

Abstract: A computer-implemented method for training a multi-task neural network for predicting a plurality of T tasks, T?2, simultaneously based on input data, the method comprising: a) providing a multi-task neural network, a training dataset for training the neural network on the plurality of T tasks, and a validation dataset ? for validating the neural network on the plurality of T tasks; and b) training the multi-task neural network for the plurality of T tasks across a predefined number Nepoch of training epochs by using the training dataset and the validation dataset ? such that a combined loss function is minimized within the predefined number Nepoch of training epochs.

Abstract: A portable hemopurification system can remove one or more uremic toxins from blood of a patient. The hemopurification system includes a dialyzer with a dialyzing membrane. Blood from a patient is provided to the dialyzer on one side of the dialyzing membrane, and a negative-pressure pump applies negative pressure on the opposite side of the dialyzing membrane to provide ultrafiltration of uremic toxins from the blood. The system includes a replacement fluid line to deliver replacement fluid to the blood. The system includes load cells to measure the replacement fluid flow, effluent flow, and returned blood flow. The systems uses the fluid weights for the control of pumps to provide stable flow through the system.

Abstract: The present disclosure relates to computer-implemented methods for automatically controlling a machine includes: receiving data generated using at least one sensor of a machine; performing one or more prediction tasks on the data using a neural network, wherein the neural network includes at least one parameter tensor having at least one element, and the at least one parameter tensor was over-parameterized during training into a plurality of component tensors; and controlling the machine based on results of the one or more prediction tasks. The present disclosure further relates to a computing system for carrying out the method, a method for generating a machine learned neural network for the method, a data structure, a machine, a mobile agent, a data processing system, and a computer program, machine-readable storage medium, or a data carrier signal.

Abstract: An electronic device may have a display. The display may have light-emitting diode pixels or other pixels. Due to aging-induced degradation of the pixels, the output intensity and color of the pixels may vary over the lifetime of the display. Control circuitry in the device may use an ambient light sensor to measure ambient light and stray display light. The display light measurements may be used in monitoring aging effects so that the control circuitry can compensate for aging-induced pixel degradation. Display light measurements may be made over a sequence of image frames. The image frames in the sequence of frames may include white and black frames or frames of different colors interleaved with black frames. Different frame intensities may be used during measurements of stray display light to gather gamma curves for the display.

Abstract: A computer-implemented method includes receiving data generated using at least one sensor of a vehicle; and simultaneously performing multiple different prediction tasks on the data using a multi-task neural network, wherein the multi-task neural network comprises at least one shared parameter inference matrix comprising parameters shared between the multiple different prediction tasks, and the at least one shared parameter inference matrix was over-parameterized during training into at least one shared parameter matrix and multiple task-specific parameter matrices, each of the multiple task-specific parameter matrices being associated with a different one of the multiple different tasks.

Abstract: An optical element includes a single transparent substrate. A first metasurface disposed on the single transparent substrate is configured to focus an input beam of optical radiation that is incident on the optical element. A second metasurface disposed on the single transparent substrate is configured to split the input beam into an array of multiple output beams.

Abstract: An electronic device may have a display. The display may have light-emitting diode pixels or other pixels. Due to aging-induced degradation of the pixels, the output intensity and color of the pixels may vary over the lifetime of the display. Control circuitry in the device may use an ambient light sensor to measure ambient light and stray display light. The display light measurements may be used in monitoring aging effects so that the control circuitry can compensate for aging-induced pixel degradation. Display light measurements may be made over a sequence of image frames. The image frames in the sequence of frames may include white and black frames or frames of different colors interleaved with black frames. Different frame intensities may be used during measurements of stray display light to gather gamma curves for the display.

Abstract: Embodiments disclosed herein generally relate to a method for monitoring optical power in a HAMR device. In one embodiment, the method includes enhancing a thermal sensor bandwidth through advanced electrical detection techniques. The advanced electrical detection techniques include obtaining calibration waveform data for a thermal sensor by calibrating the thermal sensor, obtaining real-time waveform data for the thermal sensor that may deviate from the calibration waveform data, updating the calibration waveform data to include the real-time waveform data, repeating obtaining real-time waveform data and updating the calibration waveform data during writing operations. By updating the calibration waveform data, the bandwidth of the thermal sensor is determined by a fixed sampling time interval, and the thermal sensor rise time to steady state would not be a limitation to its response time.

Abstract: Embodiments disclosed herein generally relate to a method for monitoring optical power in a HAMR device. In one embodiment, the method includes enhancing a thermal sensor bandwidth through advanced electrical detection techniques. The advanced electrical detection techniques include obtaining calibration waveform data for a thermal sensor by calibrating the thermal sensor, obtaining real-time waveform data for the thermal sensor that may deviate from the calibration waveform data, updating the calibration waveform data to include the real-time waveform data, repeating obtaining real-time waveform data and updating the calibration waveform data during writing operations. By updating the calibration waveform data, the bandwidth of the thermal sensor is determined by a fixed sampling time interval, and the thermal sensor rise time to steady state would not be a limitation to its response time.

Abstract: Embodiments disclosed herein generally relate to a method for monitoring optical power in a HAMR device. In one embodiment, the method includes enhancing a thermal sensor bandwidth through advanced electrical detection techniques. The advanced electrical detection techniques include obtaining calibration waveform data for a thermal sensor by calibrating the thermal sensor, obtaining real-time waveform data for the thermal sensor that may deviate from the calibration waveform data, updating the calibration waveform data to include the real-time waveform data, repeating obtaining real-time waveform data and updating the calibration waveform data during writing operations. By updating the calibration waveform data, the bandwidth of the thermal sensor is determined by a fixed sampling time interval, and the thermal sensor rise time to steady state would not be a limitation to its response time.

Abstract: Embodiments disclosed herein generally relate to a method for monitoring optical power in a HAMR device. In one embodiment, the method includes enhancing a thermal sensor bandwidth through advanced electrical detection techniques. The advanced electrical detection techniques include obtaining calibration waveform data for a thermal sensor by calibrating the thermal sensor, obtaining real-time waveform data for the thermal sensor that may deviate from the calibration waveform data, updating the calibration waveform data to include the real-time waveform data, repeating obtaining real-time waveform data and updating the calibration waveform data during writing operations. By updating the calibration waveform data, the bandwidth of the thermal sensor is determined by a fixed sampling time interval, and the thermal sensor rise time to steady state would not be a limitation to its response time.

Abstract: A system, according to one embodiment, includes a magnetic head having: a near field transducer, an optical waveguide for illumination of the near field transducer, and an anti-reflection block positioned along the optical waveguide farther from a media facing side of the magnetic head than the near field transducer. The anti-reflection block is positioned a distance from the near field transducer to destructively interfere with light reflected away from the near field transducer. Other systems, methods, and computer program products are described in additional embodiments.

Abstract: A method for annealing a metal antenna of a near field optical transducer of a magnetic write element without inadvertently heat damaging the read element. A heating element is placed within a write head build in a cerf region outside of the active area of the read and write heads. A layer of thermally conductive, electrically insulating material is formed over the heating element to separate the heating element from the antenna. The thermally conductive, electrically insulating layer is preferably in contact with the antenna. A current can be applied to the heating element to heat the antenna to a temperature that is at or above the operating temperature of the optical transducer. After annealing, the heater element can be removed by the lapping process that is used to define the media facing surface of the head.

Abstract: A method for annealing a metal antenna of a near field optical transducer of a magnetic write element without inadvertently heat damaging the read element. A heating element is placed within a write head build in a cerf region outside of the active area of the read and write heads. A layer of thermally conductive, electrically insulating material is formed over the heating element to separate the heating element from the antenna. The thermally conductive, electrically insulating layer is preferably in contact with the antenna. A current can be applied to the heating element to heat the antenna to a temperature that is at or above the operating temperature of the optical transducer. After annealing, the heater element can be removed by the lapping process that is used to define the media facing surface of the head.

Abstract: In a heat-assisted magnetic recording hard disk drive, a laser module includes a submount-integrated photodetector configured to receive optical energy from a laser by way of a head slider. The submount may be formed of a semiconductor material such as a crystalline silicon material, and the photodetector may be a photodiode that is integrally formed with the submount. A HAMR head slider may comprise a feedback waveguide configured to guide optical energy from the laser through the slider to a feedback photodiode at an interface of the submount and the slider, to detect the optical energy transmitted through the slider to the slider air bearing surface (ABS). A back facet photodiode may also be integrally formed with the submount and configured to receive back facet optical energy to detect the optical energy generated by the laser.

Abstract: Embodiments disclosed herein generally relate to a HAMR device for use in a magnetic reading system. The HAMR device has a primary waveguide extending from a media facing surface to a surface opposite the media facing surface. In one embodiment, the HAMR head has a secondary waveguide having a first and second end positioned within the HAMR head. The secondary waveguide is positioned near the primary waveguide, and the first and second ends of the secondary waveguide are non-reflecting. A thermal sensor is coupled to the secondary waveguide, and responds only to optical power fluctuations in the primary waveguide. In another embodiment, the HAMR head has a second thermal sensor rather than a secondary waveguide. The first thermal sensor and the second thermal sensor are coupled together to act as one optical power sensor and are disposed on the primary waveguide.

Abstract: Embodiments disclosed herein generally relate to a HAMR device for use in a magnetic reading system. The HAMR device has a primary waveguide extending from a media facing surface to a surface opposite the media facing surface. In one embodiment, the HAMR head has a secondary waveguide having a first and second end positioned within the HAMR head. The secondary waveguide is positioned near the primary waveguide, and the first and second ends of the secondary waveguide are non-reflecting. A thermal sensor is coupled to the secondary waveguide, and responds only to optical power fluctuations in the primary waveguide. In another embodiment, the HAMR head has a second thermal sensor rather than a secondary waveguide. The first thermal sensor and the second thermal sensor are coupled together to act as one optical power sensor and are disposed on the primary waveguide.

Abstract: In one general embodiment, an apparatus includes a write pole, a near field transducer, a waveguide for delivering light to the near field transducer, and a first heating device positioned between the write pole and at least one of the waveguide and the near field transducer.

Abstract: Embodiments disclosed herein generally relate to a magnetic write head including a media facing surface and a surface opposite the media facing surface. The magnetic write head also includes a SSC having a first end extending to the surface opposite the media facing surface. The first end of the SSC has at least two tips. The multiple-tipped first end of the SSC is configured to receive a laser beam and the laser beam coupling efficiency is improved by having the multiple-tipped first end.