Abstract: Some example embodiments provide a storage device including a non-volatile memory; a microcontroller unit configured to monitor statuses of constituent elements of the storage device and output an abnormality signal based on the constituent elements not being in a normal status; and a storage controller configured to perform a first check and a second check different from the first check on a first firmware code stored in the microcontroller unit in response to the abnormality signal and perform recovery by using a second firmware code stored in the non-volatile memory based on at least one of the first check and the second check having failed.

Abstract: Disclosed are methods of operating a storage device which communicates with a host device includes receiving a first power supply voltage and a second power supply voltage from the host device, executing, by a micro controller unit (MCU) of the storage device, a first power-on sequence for sequentially generating a plurality of internal voltages of the storage device based on the second power supply voltage, detecting, by the MCU, a first power failure corresponding to the first power-on sequence, and performing, by the MCU, a second power-on sequence for sequentially generating the plurality of internal voltages in response to detecting the first power failure.

Abstract: A method of programming an activation function is provided. The method includes generating a segment data for segmenting the activation function; segmenting the activation function into a plurality of segments using the segment data; and approximating at least one segment of the plurality of segments to a programmable segment.

Abstract: A method for operating a storage device according to some example embodiments may include monitoring a power supply voltage provided by an external host device; detecting an error of the power supply voltage; generating self-monitoring data of the power supply voltage, and requesting real-time state information to the external host device, based on detection of the error of the power supply voltage; receiving the real-time state information from the external host device; generating result data for indicating whether the error of the power supply voltage is caused by the storage device based on the self-monitoring data and the real-time state information; and transmitting the result data to the external host device.

Abstract: An activation function conversion program unit and method may be configured to approximate a target activation function to a programmed activation function through machine-learning of an artificial neural network. The method may include setting up a target activation function; approximating the target activation function to a programmed activation function by machine-learning an artificial neural network; and converting the programmed activation function into a slope and offset and storing it in a lookup table. Accordingly, the computation speed and power consumption of the programmed activation function execution unit of an NPU may be optimized.

Abstract: A storage device includes a plurality of power circuits configured to operate sequentially to drive the storage device, a control circuit configured to, monitor a plurality of first monitoring signals received from the plurality of power circuits, and generate a retry request in response to detection of hardware initialization failure in at least one of the plurality of power circuits based on the plurality of first monitoring signals, and a power sequence circuit configured to, monitor a plurality of second monitoring signals received from the plurality of power circuits and the control circuit, and retry a hardware initialization operation for at least one power circuit of the plurality of power circuits based on the retry request and the plurality of second monitoring signals.

Abstract: A method for controlling an electronic device is provided.

Abstract: An activation function conversion program unit and method may be configured to approximate a target activation function to a programmed activation function through machine-learning of an artificial neural network. The method may include setting up a target activation function; approximating the target activation function to a programmed activation function by machine-learning an artificial neural network; and converting the programmed activation function into a slope and offset and storing it in a lookup table. Accordingly, the computation speed and power consumption of the programmed activation function execution unit of an NPU may be optimized.

Abstract: A method for selecting one or more items by an electronic device is provided.

Abstract: A method of programming an activation function is provided. The method includes generating a segment data for segmenting the activation function; segmenting the activation function into a plurality of segments using the segment data; and approximating at least one segment of the plurality of segments to a programmable segment.

Abstract: A storage device includes a plurality of power circuits configured to operate sequentially to drive the storage device, a control circuit configured to, monitor a plurality of first monitoring signals received from the plurality of power circuits, and generate a retry request in response to detection of hardware initialization failure in at least one of the plurality of power circuits based on the plurality of first monitoring signals, and a power sequence circuit configured to, monitor a plurality of second monitoring signals received from the plurality of power circuits and the control circuit, and retry a hardware initialization operation for at least one power circuit of the plurality of power circuits based on the retry request and the plurality of second monitoring signals.

Abstract: Some example embodiments provide a storage device including a non-volatile memory; a microcontroller unit configured to monitor statuses of constituent elements of the storage device and output an abnormality signal based on the constituent elements not being in a normal status; and a storage controller configured to perform a first check and a second check different from the first check on a first firmware code stored in the microcontroller unit in response to the abnormality signal and perform recovery by using a second firmware code stored in the non-volatile memory based on at least one of the first check and the second check having failed.

Abstract: An optical bio-sensing device includes a transparent substrate covering a top of a space accommodating therein a sample containing a target bio-material; a signal converter fixed to the transparent substrate, and including the upconversion nanoparticles for receiving incident light and emitting converted light of a wavelength shorter than a wavelength of the incident light; a signal reflector including retroreflection particles bindable to the signal converter via the target bio-material, wherein the retroreflection particles retroreflect the converted light; a light source for irradiating the incident light to the signal converter; and a light receiver for receiving light retroreflected from the signal reflector.

Abstract: A method of programming an activation function is provided. The method includes generating a segment data for segmenting the activation function; segmenting the activation function into a plurality of segments using the segment data; and approximating at least one segment of the plurality of segments to a programmable segment.

Abstract: An optical gene biosensor is disclosed. The optical gene biosensor includes a substrate; a molecular beacon anchored to the substrate, wherein the molecular beacon includes an oligonucleotide specifically binding to a target nucleic acid molecule and a first compound bound to a first terminal of the oligonucleotide; an optical marker specifically binding to the first compound, wherein the optical marker is configured to retro-reflect irradiated light; a light source for irradiating the optical marker with light; and a light-receiver for receiving light retro-reflected from the optical marker. The optical gene biosensor may perform accurate quantitative analysis of a target nucleic acid molecule using both non-spectral and spectral light sources.

Abstract: A method for programming an activation function is provided. The method includes generating segment data for segmenting the activation function; segmenting the activation function into a plurality of segments using the segment data; and approximating at least one segment of the plurality of segments as a programmable segment. An apparatus for performing the method may include a programmable activation function generator configured to generate segment data for segmenting an activation function; segment the activation function into a plurality of segments using the generated segment data; and approximate at least one segment of the plurality of segments as a programmable segment. By using segment data, various non-linear activation functions, particularly newly proposed or known activation functions with some modifications, can be programmed to be processable in hardware.

Abstract: A method for selecting one or more items by an electronic device is provided.

Abstract: A method for selecting one or more items by an electronic device is provided. The method includes receiving a first input for selecting two more points on a screen, in response to the first input, executing a multi-selection mode for selecting one or more items, receiving a second input which is inputted in succession to the first input, and in response to the second input, selecting the one or more items.

Abstract: A method for programming an activation function is provided. The method includes generating segment data for segmenting the activation function; segmenting the activation function into a plurality of segments using the segment data; and approximating at least one segment of the plurality of segments as a programmable segment. An apparatus for performing the method may include a programmable activation function generator configured to generate segment data for segmenting an activation function; segment the activation function into a plurality of segments using the generated segment data; and approximate at least one segment of the plurality of segments as a programmable segment. By using segment data, various non-linear activation functions, particularly newly proposed or known activation functions with some modifications, can be programmed to be processable in hardware.

Abstract: Computer-readable media and related methods for semantic slide autolayouts are disclosed. A processor is configured to display a default layout of a presentation slide within a graphical user interface (GUI). The default layout includes a text field and an asset field. The text field is configured to receive text to be displayed in the presentation slide, and the asset field is configured to receive digital assets to be displayed in the presentation slide. The processor is also configured to automatically determine an intent to emphasize the text or the digital assets within the presentation slide, and automatically reformat the presentation slide to emphasize the text or the digital assets based on the determined intent.