Abstract: There is provided an information processing device to improve the accuracy of estimation using a student network, the information processing device including an estimation unit that estimates an object class of an object included in an input image using a student network generated based on a teacher network generated by machine learning using images stored in a large-scale image database as training data. The student network is generated by machine learning using, as training data, synthetic images obtained using the teacher network and real environment images acquired by a plurality of different modalities in a real environment in which estimation by the estimation unit is expected to be executed.

Abstract: An apparatus, method and computer program is described for detecting a pre-event signature based on sensor data, wherein the pre-event signature is indicative of a prediction of a first event; determining an event masking action corresponding to the first event; and causing performance, in response to the detection of the pre-event signature, of the event masking action for masking the first event.

Abstract: The invention proposes systems and methods for testing on-device machine learning models such as network models of a device based on an access to normal input and output channels, wherein a structure of a model is designed through training such that the model produces distinctive outputs to a given set of test inputs only so long as its internal structure remains in a desired state. Models may be rendered susceptible to such testing via model pre-training with training inputs designed to train the model into an appropriate structure and/or to cause the model to produce a distinctive output if later presented with a certain set of test inputs. If the structure of the model remains within allowable limits, the device will produce a predictable output when tested. If not, the device may be enforced to return to a pre-trained model.

Abstract: A terminal apparatus (20) according to an embodiment includes: a first ranging unit (203) including a light source unit (2110) and a light reception unit (2111) and configured to perform range-finding on a basis of light emitted from the light source unit and light received by the light reception unit, and a control unit (219) configured to control the first ranging unit, in which the control unit controls light emission from the light source unit included in the first ranging unit in response to transmission data to transmit the transmission data to another equipment using light and to cause the first ranging unit to execute the range-finding at least once before completing transmission of the transmission data.

Abstract: Systems and methods are proposed for verifying whether a digital twin of abiological asset (or biological system) of a subject (110) indeed relates to the subject (110). Such proposals may be based on obtaining: a first signal comprising a value of a physiological parameter of the subject (e.g. determined by a physical and/or virtual sensor arrangement); and a second signal generated by the digital twin that simulates a value of a physiological parameter of the subject (110). By comparing the first and second signals, a verification result may then be determined.

Abstract: According to an example aspect of the present invention, there is provided an apparatus comprising a memory configured to store a sensor management dataset, and at least one processing core, configured to process an instruction to limit an accuracy of a first sensor in accordance with a resolution requirement of a property measured by the first sensor, determine, using the sensor management dataset, at least one second sensor which is capable of estimating the property, and cause lowering of resolution of data produced by the at least one second sensor, the extent of resolution lowering being determined based on first resolution requirement and the sensor management dataset.

Abstract: A provided method is for configuring settings of a computing device for providing more efficient and reliable acquisition of data for use in updating a personalized digital model (digital twin) of a subject. The method comprises configuring settings of a biometric authentication function of a computing device so as to provide for overlap in the input data requirements of the biometric authentication function and the input data requirements of the digital twin. There may be different authentication protocols available at the computing device, each requiring different input sensing data. Based on knowledge of these different authentication protocols and data requirements, and based on knowledge of data input needs of a digital twin, an authentication protocol can be selected and/or its settings adjusted, so that when performing biometric authentication, the same acquired sensor data can also be used for deriving physiological parameter data of the subject, for updating the digital twin.

Abstract: Using a transformation based at least in part on a non-simple orthogonal or unitary matrix, data may be transmitted over a data bus in a manner that is resilient to one or more types of signal noise, that does not require a common reference at the transmission and acquisition points, and/or that has a pin-efficiency that is greater than 50% and may approach that of single-ended signaling. Such transformations may be implemented in hardware in an efficient manner. Hybrid transformers that apply such transformations to selected subsets of signals to be transmitted may be used to adapt to various signal set sizes and/or transmission environment properties including noise and physical space requirements of given transmission environments.

Abstract: Acquisition of a three-dimensional shape of an object by a user is efficiently supported. There is provided an information processing device including a control unit that controls presentation of feedback related to arrangement of a plurality of mirror surfaces arranged around an object to be imaged and used for acquiring a three-dimensional shape of the object. Furthermore, there is provided an information processing method including controlling, by a processor, presentation of feedback related to arrangement of a plurality of mirror surfaces arranged around an object to be imaged and used for acquiring a three-dimensional shape of the object.

Abstract: Provided is an information processing apparatus (500) including a control section (504) that dynamically changes each parameter related to display of a virtual object, the parameter controlling display of the virtual object on each display device according to a method of expressing an image, the method being assigned for display of the image to each of a plurality of display devices that display images related to the same virtual object.

Abstract: Using a transformation based at least in part on a non-simple orthogonal or unitary matrix, data may be transmitted over a data bus in a manner that is resilient to one or more types of signal noise, that does not require a common reference at the transmission and acquisition points, and/or that has a pin-efficiency that is greater than 50% and may approach that of single-ended signaling. Such transformations may be implemented in hardware in an efficient manner. Hybrid transformers that apply such transformations to selected subsets of signals to be transmitted may be used to adapt to various signal set sizes and/or transmission environment properties including noise and physical space requirements of given transmission environments.

Abstract: There is provided an information processing device to improve the accuracy of estimation using a student network, the information processing device including an estimation unit that estimates an object class of an object included in an input image using a student network generated based on a teacher network generated by machine learning using images stored in a large-scale image database as training data. The student network is generated by machine learning using, as training data, synthetic images obtained using the teacher network and real environment images acquired by a plurality of different modalities in a real environment in which estimation by the estimation unit is expected to be executed.

Abstract: Systems and methods are proposed for training and employing a machine learning algorithm for identifying uncertain Digital Twin model parameters. Such proposals may make use of a set of DTs that is chosen such that it comprises DTs that have achieved a target/predetermined morbidity scenario. The training may then identify uncertain model parameters for the set of DTs. Using the identified uncertain model parameters, the set of DTs may then be run so as to assess whether the target morbidity scenario is achieved.

Abstract: Using a transformation based at least in part on a non-simple orthogonal or unitary matrix, data may be transmitted over a data bus in a manner that is resilient to one or more types of signal noise, that does not require a common reference at the transmission and acquisition points, and/or that has a pin-efficiency that is greater than 50% and may approach that of single-ended signaling. Such transformations may be implemented in hardware in an efficient manner. Hybrid transformers that apply such transformations to selected subsets of signals to be transmitted may be used to adapt to various signal set sizes and/or transmission environment properties including noise and physical space requirements of given transmission environments.

Abstract: A method and apparatus for determining one or more first devices that are Internet devices meeting all of the following conditions: residing at a given location; equipped with one or more ambience sensing capable sensors; and operation mode being such that their ambience sensing capable sensors should not cause transmission of data. One or more second devices are determined that are Internet devices at the given location and equipped with one or more elements capable of causing an ambient stimulation detectable by the sensors of one or more first devices. Data transmissions of the first devices are monitored. Issuing of the ambient stimulation is caused by a subset of the one or more second devices. It is determined whether the issuing of the ambient stimulation caused a significant change in the monitored data transmissions of the first devices.

Abstract: An apparatus, method and computer program is described for detecting a pre-event signature based on sensor data, wherein the pre-event signature is indicative of a prediction of a first event; determining an event masking action corresponding to the first event; and causing performance, in response to the detection of the pre-event signature, of the event masking action for masking the first event.

Abstract: According to an example aspect of the present invention, there is provided an apparatus comprising at least one processing core, at least one memory including computer program code, the at least one memory and the computer program code being configured to, with the at least one processing core, cause the apparatus at least to receive from a pressure sensor matrix data describing a time-varying pressure interaction of the pressure sensor matrix with a second pressure sensor matrix, process the data to obtain a bit sequence, and use the bit sequence as a shared secret in a cryptographic procedure with a device.

Abstract: A device for accessing a distributed ledger is provided. The distributed ledger is maintained by a plurality of virtual nodes. The device has a circuitry, which is configured to access the distributed ledger through at least one virtual node of the plurality of virtual nodes.

Abstract: An apparatus for providing a virtual device has a circuitry. The circuitry searches for resource devices providing characteristics of a set of characteristics of the virtual device in a distributed ledger. The distributed ledger includes information about multiple resource devices. The circuitry provides the virtual device by selecting resource devices providing characteristics of the set of characteristics of the virtual device.

Abstract: Using a transformation based at least in part on a non-simple orthogonal or unitary matrix, data may be transmitted over a data bus in a manner that is resilient to one or more types of signal noise, that does not require a common reference at the transmission and acquisition points, and/or that has a pin-efficiency that is greater than 50% and may approach that of single-ended signaling. Such transformations may be implemented in hardware in an efficient manner. Hybrid transformers that apply such transformations to selected subsets of signals to be transmitted may be used to adapt to various signal set sizes and/or transmission environment properties including noise and physical space requirements of given transmission environments.