Abstract: A data processing circuit and a fault-mitigating method are provided. In the method, multiple sub-sequences are divided from sequence data. A first sub-sequence of the sub-sequences is accessed from a memory for a multiply-accumulate (MAC) operation to obtain a first computed result. The MAC operation is performed on a second sub-sequence of the sub-sequences in the memory to obtain a second computed result. The first and the second computed results are combined, where the combined result of the first and the second computed results is related to the result of the MAC operation on the sequence data directly. Accordingly, the error rate could be reduced, so as to mitigate fault.

Abstract: A data processing circuit and a fault mitigating method are provided. The method is adapted for a memory having at least one fault bit. The memory provides a block for data storage. A difference between an output of a value of a plurality of bits input to at least one computing layer in a neural network and a correct value is determined. The bits are respectively considered the at least one fault bit. A repair condition is determined based on the difference. The repair condition includes a correspondence between a position where the fault bit is located in the block and at least one non-fault bit in the memory. A value of at least one non-fault bit of the memory replaces a value of the fault bit based on the repair condition.

Abstract: A chip package includes an application chip, a micro-electromechanical systems (MEMS) chip, a conductive element, a bonding wire, and a molding compound. The application chip has a conductive pad. The MEMS chip is located on the application chip, and includes a main body and a cap. The main body is located between the cap and the application chip. The main body has a conductive pad. The conductive element is located on the conductive pad of the main body of the MEMS chip. The bonding wire extends from the conductive element to the conductive pad of the application chip. The molding compound is located on the application chip and surrounds the MEMS chip. The conductive element and the bonding wire are located in the molding compound.

Abstract: A data processing circuit and a fault-mitigating method are provided. A first data is written into a memory. A computed result is determined according to one or more adjacent bits of the first data at faulty bits. According to the computed result, new values are determined. The new values replace the values of the first data at the faulty bits to form a second data. The first data includes multiple bits. The first data is image-related data, weights used by a multiply-accumulate (MAC) for extracting features of images, and/or values used by an activation calculation. The adjacent bits are adjacent to the faulty bits. The computed result is obtained through computing the values of the first data at non-faulty bits of the memory. Accordingly, an influence of a memory fault is reduced.

Abstract: A memory apparatus and a data rearrangement method for computing in memory (CIM) are provided. The method includes determining whether first sequence data has two target bits that are both of a first value, inserting a non-target bit of a second value between the two target bits that are both of the first value and adjacent to each other to generate second sequence data, and receiving the second sequence data through memory cells in a memory to perform a multiply-accumulate (MAC) operation on the second sequence data. Each bit in the first sequence data is the first value or the second value. One of the two target bits is located adjacent to the other one of the two target bits in the first sequence data. The two target bits and the non-target bit are located in the first sequence data. Accordingly, the error rate is decreased.

Abstract: An error calibration apparatus and method are provided. The method is adapted for calibrating a machine learning (ML) accelerator. The ML accelerator achieves computation by using an analog circuit. An error between an output value of one or more computing layers of a neural network and a corresponding corrected value is determined. The computation of the computing layers is achieved by the analog circuit. A calibration node is generated according to the error. The calibration node is located at the next layer of the computing layers. The calibration node is used to minimize the error. The calibration node is achieved by a digital circuit. Accordingly, error and distortion of the analog circuit could be reduced.

Abstract: An error calibration apparatus and method are provided. The method is adapted for calibrating a machine learning (ML) accelerator. The ML accelerator achieves computation by using an analog circuit. An error between an output value of one or more computing layers of a neural network and a corresponding corrected value is determined. The computation of the computing layers is achieved by the analog circuit. A calibration node is generated according to the error. The calibration node is located at the next layer of the computing layers. The calibration node is used to minimize the error. The calibration node is achieved by a digital circuit. Accordingly, error and distortion of the analog circuit could be reduced.

Abstract: A data processing circuit and a fault mitigating method are provided. The method is adapted for a memory having at least one fault bit. The memory provides a block for data storage. A difference between an output of a value of a plurality of bits input to at least one computing layer in a neural network and a correct value is determined. The bits are respectively considered the at least one fault bit. A repair condition is determined based on the difference. The repair condition includes a correspondence between a position where the fault bit is located in the block and at least one non-fault bit in the memory. A value of at least one non-fault bit of the memory replaces a value of the fault bit based on the repair condition.

Abstract: A data processing circuit and a fault-mitigating method are provided. In the method, multiple sub-sequences are divided from sequence data. A first sub-sequence of the sub-sequences is accessed from a memory for a multiply-accumulate (MAC) operation to obtain a first computed result. The MAC operation is performed on a second sub-sequence of the sub-sequences in the memory to obtain a second computed result. The first and the second computed results are combined, where the combined result of the first and the second computed results is related to the result of the MAC operation on the sequence data directly. Accordingly, the error rate could be reduced, so as to mitigate fault.

Abstract: A data processing circuit and a fault-mitigating method, which are adapted for a memory having a faulty bit, are provided. The memory is configured to store data related to an image, a weight for a multiply-accumulate (MAC) operation of image feature extraction, and/or a value for an activation operation. Sequence data is written into the memory. The bit number of the sequence data equals to the bit number used for storing data in a sequence block of the memory. The sequence data is accessed from the memory, wherein the access of the faulty bit in the memory is ignored. The value of the faulty bit is replaced by the value of a non-faulty bit in the memory to form new sequence data. The new sequence data is used for MAC. Accordingly, the accuracy of image recognition can be improved for the faulty memory.

Abstract: A data processing circuit and a fault-mitigating method, which are adapted for a memory having a faulty bit, are provided. The memory is configured to store data related to an image, a weight for a multiply-accumulate (MAC) operation of image feature extraction, and/or a value for an activation operation. Sequence data is written into the memory. The bit number of the sequence data equals to the bit number used for storing data in a sequence block of the memory. The sequence data is accessed from the memory, wherein the access of the faulty bit in the memory is ignored. The value of the faulty bit is replaced by the value of a non-faulty bit in the memory to form new sequence data. The new sequence data is used for MAC. Accordingly, the accuracy of image recognition can be improved for the faulty memory.

Abstract: Access control device includes a first electronic device and a second electronic device. The first electronic device includes a storing module and a processing module. The storing module stores an information about the visitor, and the processing module processes the information to generate an identification figure. The second electronic device includes an executing module, and the executing module determines an access permission about the visitor for entering into a restricted zone, according to the identification figure. The disclosure further provides an access control method.

Abstract: An electronic device includes a base member and a display member rotatably coupled to the base member. A keyboard and a touchpad are located on a working surface of the base member. The touchpad includes a middle touch area, a left touch area, and a right touch area. When the touchpad works in a full touch mode, the middle touch area, the left touch area, and the right touch area are all enabled to sense and recognize touch gestures input by a user of the electronic device. When the touchpad works in a partial touch mode, at least one of the middle touch area, the left touch area, and the right touch area is disabled from sensing and recognizing touch gestures.

Abstract: An electronic device includes a cover member rotatably connected to a base member. A magnetic lock is located on a working surface of the base member. A touch-sensitive bar is located on a side surface of the base member. When the cover member and the base member are closed, the magnetic lock provides a magnetic attraction between the cover member and the base member. When the touch-sensitive bar detects a touch gesture input by a user and the touch gesture matches a predefined touch gesture, the magnetic lock releases the magnetic attraction between the cover member and the base member.

Abstract: An electronic device includes a front panel, a display located on the front panel, a rear panel opposite to the front panel, a camera lens located on at least one of the front panel and the rear panel, a side flange connected between the front panel and the rear panel, and a touch strip is located on the side flange. The touch panel is configured to receive a control operation to adjust an image capturing function.

Abstract: An electronic device includes a base and a cover rotatably attached to the base. An identification unit is located on the cover. The identification unit is used to sense a gesture to rotate the cover relative to the base. The disclosure further offers a control method for the electronic device.

Abstract: An electronic device includes a read only memory (ROM), a processing unit coupled to the ROM, a first sensing area coupled to the processing unit and a second sensing area coupled to the processing unit. The processing unit starts up when the first sensing area and the second sensing area are covered simultaneously by fingers of a hand of a user of the electronic device in a predetermined period of time.

Abstract: An electronic device includes a touch-sensitive screen, a cursor locating module, a touch detecting module, and a user interface (UI) generating module. The cursor locating module locates a cursor in a graphical user interface (GUI) displayed on the touch-sensitive screen. The touch detecting module detects whether a point of the touch-sensitive screen corresponding to the cursor is continuously pressed for a preset time duration. When the point is continuously pressed for the preset time duration, the UI generating module generates a search UI and displays the search UI adjacent to the cursor in the GUI. A method for searching data in a touch-sensitive device is also provided.