Abstract: Many unsupervised domain adaptation (UDA) methods have been proposed to bridge the domain gap by utilizing domain invariant information. Most approaches have chosen depth as such information and achieved remarkable successes. Despite their effectiveness, using depth as domain invariant information in UDA tasks may lead to multiple issues, such as excessively high extraction costs and difficulties in achieving a reliable prediction quality. As a result, we introduce Edge Learning based Domain Adaptation (ELDA), a framework which incorporates edge information into its training process to serve as a type of domain invariant information. Our experiments quantitatively and qualitatively demonstrate that the incorporation of edge information is indeed beneficial and effective, and enables ELDA to outperform the contemporary state-of-the-art methods on two commonly adopted benchmarks for semantic segmentation based UDA tasks.

Abstract: Methods, systems and apparatus for memory devices with multiple string select line (SSL) cuts are provided. In one aspect, a semiconductor device includes: a three-dimensional (3D) array of memory cells and a plurality of common source lines (CSLs) configured to separate the 3D array of memory cells into a plurality of portions. Each portion of the plurality of portions is between two adjacent CSLs and includes a plurality of conductive layers separated from each other by insulating layers and a plurality of vertical channels arranged orthogonally through the plurality of conductive layers and the insulating layers, each of the plurality of vertical channels including a string of memory cells. A top part of each portion of one or more portions includes at least two SSL cuts configured to separate the portion into multiple independent units, and each of the independent units is selectable by a corresponding SSL of multiple SSLs.

Abstract: A pressure sensor includes a first electrode, a plurality of cavities, and a second electrode. The second electrode is disposed opposite the first electrode through the plurality of cavities. The second electrode includes a flat structure spanning two adjacent cavities of the plurality of cavities.

Abstract: A pressure sensor comprises a polysilicon sensing membrane. The pressure sensor further includes one or more polysilicon electrodes disposed over a silicon substrate. The sensor also includes one or more polysilicon routing layers that electrically connects electrodes of the one or more polysilicon electrodes to one another, wherein the polysilicon sensing membrane deforms responsive to a stimuli and changes a capacitance between the polysilicon sensing membrane and the one or more polysilicon electrodes. The sensor also includes one or more vacuum cavities positioned between the polysilicon sensing membrane and the one or more polysilicon electrodes.

Abstract: A method includes depositing a passivation layer on a substrate; depositing and patterning a first polysilicon layer on the passivation layer; depositing and patterning a first oxide layer on the first polysilicon layer forming a patterned first oxide layer; depositing and patterning a second polysilicon layer on the patterned first oxide layer. A portion of the second polysilicon layer directly contacts a portion of the first polysilicon layer. A portion of the patterned second polysilicon layer corresponds to a bottom electrode. A second oxide layer is deposited on the patterned second polysilicon layer and on an exposed portion of the patterned first oxide layer. A portion of the second oxide layer corresponding to a sensing cavity is etched, exposing the bottom electrode. Another substrate is bonded to the second oxide layer enclosing the sensing cavity. A top electrode is disposed within the another substrate and positioned over the bottom electrode.

Abstract: A device includes a microelectromechanical system (MEMS) sensor die comprising a deformable membrane, a MEMS heating element, and a substrate. The MEMS heating element is integrated within a same layer and a same plane as the deformable membrane. The MEMS heating element surrounds the deformable membrane and is separated from the deformable membrane through a trench. The MEMS heating element is configured to generate heat to heat up the deformable membrane. The substrate is coupled to the deformable membrane.

Abstract: A device includes a microelectromechanical system (MEMS) sensor die comprising a deformable membrane, a MEMS heating element, and a substrate. The MEMS heating element is integrated within a same layer and a same plane as the deformable membrane. The MEMS heating element surrounds the deformable membrane and is separated from the deformable membrane through a trench. The MEMS heating element is configured to generate heat to heat up the deformable membrane. The substrate is coupled to the deformable membrane.

Abstract: The present disclosure provides an aerosol sensing method. The aerosol sensing method includes steps of providing an entering process, providing a particle collecting process and providing a measuring process. The entering process is to allow an aerosol to enter a chamber of a thermal-piezoresistive oscillator-based aerosol sensor, and a thermal-piezoresistive resonator is disposed in the chamber. The particle collecting process is to allow particulate matters in the aerosol to land on at least one proof-mass of the thermal-piezoresistive resonator when the thermal-piezoresistive resonator is not driven. The measuring process is to use an electrical signal to drive the thermal-piezoresistive resonator and measure a resonant frequency of the thermal-piezoresistive resonator. The particle collecting process and the measuring process are operated in a repetitive cycle for measuring changes of the resonant frequency of the thermal-piezoresistive resonator to measure the particulate matters of the aerosol.

Abstract: A device includes a microelectromechanical system (MEMS) sensor die comprising a deformable membrane, a MEMS heating element, and a substrate. The MEMS heating element is integrated within a same layer and a same plane as the deformable membrane. The MEMS heating element surrounds the deformable membrane and is separated from the deformable membrane through a trench. The MEMS heating element is configured to generate heat to heat up the deformable membrane. The substrate is coupled to the deformable membrane.

Abstract: The present disclosure provides an aerosol sensing method. The aerosol sensing method includes steps of providing an entering process, providing a particle collecting process and providing a measuring process. The entering process is to allow an aerosol to enter a chamber of a thermal-piezoresistive oscillator-based aerosol sensor, and a thermal-piezoresistive resonator is disposed in the chamber. The particle collecting process is to allow particulate matters in the aerosol to land on at least one proof-mass of the thermal-piezoresistive resonator when the thermal-piezoresistive resonator is not driven. The measuring process is to use an electrical signal to drive the thermal-piezoresistive resonator and measure a resonant frequency of the thermal-piezoresistive resonator. The particle collecting process and the measuring process are operated in a repetitive cycle for measuring changes of the resonant frequency of the thermal-piezoresistive resonator to measure the particulate matters of the aerosol.

Abstract: A nucleic acid construct comprising a genetic engineered heterogeneous nuclear ribonucleoprotein (hnRNP) A1 gene is provided. A transgenic mouse in which the expression of hnRNP A1 gene has been disrupted is also provided. The mouse is useful for studying the role of hnRNP A1 gene in normal and disease states of a developmental disorder and muscular diseases. Therefore, a method of screening a compound for potential use in prevention and/or treatment of developmental disorder and muscular diseases is further provided.

Abstract: A nucleic acid construct comprising a genetic engineered heterogeneous nuclear ribonucleoprotein (hnRNP) A1 gene is provided. A transgenic mouse in which the expression of hnRNP A1 gene has been disrupted is also provided. The mouse is useful for studying the role of hnRNP A1 gene in normal and disease states of a developmental disorder and muscular diseases. Therefore, a method of screening a compound for potential use in prevention and/or treatment of developmental disorder and muscular diseases is further provided.

Abstract: A nucleic acid construct comprising a genetic engineered heterogeneous nuclear ribonucleoprotein (hnRNP) A1 gene is provided. A transgenic mouse in which the expression of hnRNP A1 gene has been disrupted is also provided. The mouse is useful for studying the role of hnRNP A1 gene in normal and disease states of a neurodegenerative disease or a cancer for developing therapies to treat any of these diseases. Therefore, a method of screening a compound for potential use in prevention and/or treatment of neurodegenerative disease or cancer is further provided.