Abstract: The disclosure provides a data privacy protection method, a server device, and a client device for federated learning. A public dataset is used to perform model training on a machine learning model by a server device to generate a gradient pool including multiple first gradients. The gradient pool and the machine learning model are received by a client device. The client device uses a local dataset to perform model training on the machine learning model to obtain a second gradient. A local gradient is selected from the first gradients in the gradient pool according to the second gradient using a differential privacy algorithm by the client device. An aggregated machine learning model is generated by performing model aggregation based on the local gradient by the server device.

Abstract: A two-stage machine learning model is used to for categorization of a dataset, such as transactions. A plurality of complementary base machine learning models are used to generate initial inference results and associated measures of inference confidence from the dataset, which are collected as a meta dataset. Each of the complementary models is associated with a different part of the dataset in which it has a higher accuracy in that part than the other models. The meta dataset is provided as input to a meta machine learning model, which is trained to produce a final inference result, and a confidence score model, which is trained to produce a confidence score associated with the final inference result.

Abstract: A device includes first and second semiconductor fins, first, second, third and fourth fin sidewall spacers, and first and second epitaxy structures. The first and second fin sidewall spacers are respectively on opposite sides of the first semiconductor fin. The third and fourth fin sidewall spacers are respectively on opposite sides of the second semiconductor fin. The first and third fin sidewall spacers are between the first and second semiconductor fins and have smaller heights than the second and fourth fin sidewall spacers. The first and second epitaxy structures are respectively on the first and second semiconductor fins and merged together.

Abstract: A surveillance monitoring method is provided, which includes: executing an algorithm using a camera to perform a first inference on recognition of an obstacle and recognition of a target; tracking at least one object using the camera to generate image information; performing a second inference on recognition of the obstacle and recognition of the target using a radar; tracking the at least one object using the radar to generate radar information; fusing the image information and the radar information to obtain a first recognition result; collecting environmental information using the camera or the radar, and forming a confidence level based on the environmental information, the first inference, and the second inference; and dynamically adjusting a proportion of the image information and the radar information according to the confidence level when fusing the image information and the radar information to obtain a second recognition result.

Abstract: A trained machine learning model and a training dataset used to train the trained machine learning model can be received. Based on the training dataset, unsupervised adversarial examples can be generated. Robustness of the trained machine learning model can be determined using the generated unsupervised adversarial examples. The training dataset can be augmented with the generated unsupervised adversarial examples. The trained machine learning model can be retrained using the augmented training dataset.

Abstract: A privacy data integration method and a server are provided. The privacy data integration method includes the following steps. A first processing device and a second processing device respectively obtain a first generative model and a second generative model according to a first privacy data and a second privacy data. A server generates a first generative data and a second generative data via the first generative model and the second generative model respectively. The server integrates the first generative data and the second generative data to obtain a synthetic data.

Abstract: A privacy data integration method and a server are provided. The privacy data integration method includes the following steps. A first processing device and a second processing device respectively obtain a first generative model and a second generative model according to a first privacy data and a second privacy data. A server generates a first generative data and a second generative data via the first generative model and the second generative model respectively. The server integrates the first generative data and the second generative data to obtain a synthetic data.

Abstract: A de-identification data generation apparatus, method, and non-transitory computer readable storage medium thereof are provided. The apparatus is stored with a plurality of original records, wherein each of the records has a plurality of original values corresponding to a plurality of attributes one-to-one. The apparatus decides a plurality of attribute relations (including a user-defined attribute relation) according to the original values, wherein each attribute relation is defined by two attributes. The apparatus decides a plurality of relation groups according to the attribute relations. For each relation group, the apparatus calculates a statistical distribution of the original values corresponding to the attributes in the relation group, aggregates the statistical distribution into a plurality of sub-statistical distributions, and adds noise to each sub-statistical distribution individually.

Abstract: Systems and methods for gesture control of an interactive event using multiple wearable devices are herein disclosed as comprising, in an implementation, receiving signal data indicative of at least one gesture from a first sensor of a first wearable device and a second sensor of a second wearable device, performing pre-processing on the signal data by a computing device in communication with the first wearable device and the second wearable device, performing feature extraction on the pre-processed signal data by the computing device, and determining the at least one gesture based on the feature extracted signal data and offline training data by the computing device. In an implementation, the first sensor and the second sensor comprise one or more of a three-axis accelerometer, a six-axis accelerometer, or an electromyography sensor, and can be configured to calibrate the other.

Abstract: Systems and methods for gesture control of an interactive event using multiple wearable devices are herein disclosed as comprising, in an implementation, receiving signal data indicative of at least one gesture from a first sensor of a first wearable device and a second sensor of a second wearable device, performing pre-processing on the signal data by a computing device in communication with the first wearable device and the second wearable device, performing feature extraction on the pre-processed signal data by the computing device, and determining the at least one gesture based on the feature extracted signal data and offline training data by the computing device. In an implementation, the first sensor and the second sensor comprise one or more of a three-axis accelerometer, a six-axis accelerometer, or an electromyography sensor, and can be configured to calibrate the other.

Abstract: A radar wave sensing apparatus including a rotation element, a nanosecond pulse near-field sensor and a control unit is provided. The nanosecond pulse near-field sensor emits an incident radar wave and receives a reflection radar wave of the incident radar wave hitting on a surface of the rotation element to obtain a repetition frequency variation of the reflection radar wave corresponding to the incident radar wave. The control unit calculates a vibration of the rotation element according to the repetition frequency variation.

Abstract: A radar wave sensing apparatus including a rotation element, a nanosecond pulse near-field sensor and a control unit is provided. The nanosecond pulse near-field sensor emits an incident radar wave and receives a reflection radar wave of the incident radar wave hitting on a surface of the rotation element to obtain a repetition frequency variation of the reflection radar wave corresponding to the incident radar wave. The control unit calculates a vibration of the rotation element according to the repetition frequency variation.

Abstract: An intestinal peristalsis detecting device comprises a radio frequency (RF) signal generator, an antenna, and a mixer. After the RF signal generator sends a pulse signal, the antenna receives the pulse signal and emits a pulse electromagnetic wave to an organism. Thereafter, the antenna receives a response pulse electromagnetic wave reflected from the organism to generate a response pulse signal. The mixer couples with the RF signal generator and the antenna to mix the pulse signal and the response pulse signal.

Abstract: An Enterprise Resource Planning (ERP) system maintains a plurality of business elements. Each of the business elements represents a business function that may or may not be needed depending on a customer's business requirement. A scoping process is a process which determines what business elements are required according to the customer's business requirement and the relationships exist among some of the business elements. The ERP system provides an automatic business configuration subsystem. The automatic business configuration subsystem maintains a set of rules. Each of the rules representing a relationship between two business elements maintained at the ERP system. The automatic business configuration subsystem automatically determines which business elements are required and need to be implemented based on the set of rules. The automatic business configuration subsystem may also automatically determine whether statuses of the business elements are consistent with each other in view of the rules.

Abstract: A method for measuring physiological parameters uses statistical properties, spectrum analysis and feedback mechanisms to remove the signal noise generated by human body movement from a detected physiological signal.

Abstract: An intestinal peristalsis detecting device comprises a radio frequency (RF) signal generator, an antenna, and a mixer. After the RF signal generator sends a pulse signal, the antenna receives the pulse signal and emits a pulse electromagnetic wave to an organism. Thereafter, the antenna receives a response pulse electromagnetic wave reflected from the organism to generate a response pulse signal. The mixer couples with the RF signal generator and the antenna to mix the pulse signal and the response pulse signal.

Abstract: A method for measuring physiological parameters uses statistical properties, spectrum analysis and feedback mechanisms to remove the signal noise generated by human body movement from a detected physiological signal.

Abstract: A node disposition apparatus, a node disposition method and a computer readable medium thereof are provided. The node disposition apparatus may determine where a base station (BS) and a relay station (RS) should be disposed according to a connected mobile station amount of each node disposition candidate position in a non-transparent wireless network. Thereby, the BS and the RS may be disposed effectively to overcome the defects of the conventional BS and RS disposition way.

Abstract: A node disposition apparatus, a node disposition method and a computer readable medium thereof are provided. The node disposition apparatus may determine where a base station (BS) and a relay station (RS) should be disposed according to a connected mobile station amount of each node disposition candidate position in a non-transparent wireless network. Thereby, the BS and the RS may be disposed effectively to overcome the defects of the conventional BS and RS disposition way.

Abstract: An Enterprise Resource Planning (ERP) system maintains a plurality of business elements. Each of the business elements represents a business function that may or may not be needed depending on a customer's business requirement. A scoping process is a process which determines what business elements are required according to the customer's business requirement and the relationships exist among some of the business elements. The ERP system provides an automatic business configuration subsystem. The automatic business configuration subsystem maintains a set of rules. Each of the rules representing a relationship between two business elements maintained at the ERP system. The automatic business configuration subsystem automatically determines which business elements are required and need to be implemented based on the set of rules. The automatic business configuration subsystem may also automatically determine whether statuses of the business elements are consistent with each other in view of the rules.