公开(公告)号：US20210365789A1

公开(公告)日：2021-11-25

申请号：US17051252

申请日：2018-05-10

Applicant: Sony Corporation

Inventor： Jim RASMUSSON

IPC: G06N3/08 ,  G06N3/04

Abstract: Embodiments generally relate to training systems and methods for machine learning systems. In one embodiment the training method permits selected portions of a set of training images to be segregated by a boundary so that training image patterns in the set of training images and within the boundaries can be presented for the training while portions outside of the boundaries can be deemphasized or otherwise obscured for the training. A smooth continuous de-emphasis gradient is exercised at the boundary between the presented and the deemphasized or obscured portions of the training images.

公开(公告)号：US20200344303A1

公开(公告)日：2020-10-29

申请号：US16957077

申请日：2018-05-10

Applicant: SONY CORPORATION

Inventor： Rickard LJUNG ,  Peter ISBERG ,  Linh TRANG ,  Jim RASMUSSON

IPC: H04L29/08 ,  H04L12/26 ,  H04N19/124 ,  H04N19/14

Abstract: A sensor device operates in accordance with a method that implements a generic encoder for sensor data. The method comprises: receiving (201) input signal data of one or more sensors; encoding (202) the input signal data into formatted sensor data in a first format; and transmitting (203) the formatted sensor data to a receiving device over a communication channel. The method further comprises: configuring (205), subject to a command, said encoding to generate the formatted sensor data in a second format which differs from the first format. The command may be generated by internal analysis (206) in the sensor device, allowing the formatted sensor data to be automatically adjusted to the input signal data, or by external analysis (207) in the receiving device, allowing external control of the formatted sensor data that is transmitted by the sensor device.

公开(公告)号：US20220004886A1

公开(公告)日：2022-01-06

申请号：US17278308

申请日：2019-12-18

Applicant: Sony Corporation

Inventor： Peter EXNER ,  Jim RASMUSSON ,  Olivier MOLINER

IPC: G06N5/02 ,  G06F16/28 ,  H04L29/06

Abstract: A method (100) is disclosed, the method (100) being performed by a server device, where the server device (400) is configured to communicate with a first electronic device (300) of a plurality of electronic devices (300, 300A-300I), the first electronic device being configured to operate using a processing scheme based on a machine-learning model. The method (100) comprises receiving (S102), from the first electronic device (300), an update request for updating the machine-learning model, the update request comprising sensor data indicative of a context surrounding the first electronic device (300); generating (S107) an updated machine-learning model based on the sensor data of the update request; and transmitting (S110) an update response comprising the updated machine-learning model to the first electronic device (300).