公开(公告)号：US20240104858A1

公开(公告)日：2024-03-28

申请号：US17951185

申请日：2022-09-23

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Wen-Ling Hsu ,  Eric Zavesky ,  Louis Alexander ,  Aritra Guha ,  Jean-Francois Paiement ,  Qiong Wu ,  Zhengyi Zhou

IPC: G06T19/00 ,  A61L9/12

CPC classification number: G06T19/006 ,  A61L9/125

Abstract: Creating scent models and using scent models in cross-reality environments can include capturing experience data identifying a scent detected and a context in which the scent was detected. The experience data can be provided to a scent modeling service to generate a scent model that can represent perceived scents and perceived scent intensities for a user. The scent model can be used to generate cross-reality session data to be used in a cross-reality session presented by a cross-reality device. The cross-reality device can include a scent generator and can generate the cross-reality session using data obtained from the user device. The cross-reality device can generate a further scent during the cross-reality session based on the scent model.

公开(公告)号：US20230350977A1

公开(公告)日：2023-11-02

申请号：US17661026

申请日：2022-04-27

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Aritra Guha ,  Zhengyi Zhou ,  Balachander Krishnamurthy

IPC: G06N20/00 ,  G06K9/62

CPC classification number: G06K9/6257 ,  G06K9/6289 ,  G06N20/00

Abstract: A method performed by a processing system including at least one processor includes identifying an insufficiency in a representation of a subpopulation in training data for a machine learning model, generating simulated data to mitigate the insufficiency in the representation, and training the machine learning model using an enhanced training data set that includes the training data and the simulated data to produce a trained machine learning model. In some examples, the generating and the training may be repeated in response to determining that an output of the trained machine learning model still reflects the insufficiency in the representation of the subpopulation or reflects an insufficiency in a representation of another subpopulation. In other examples, the simulated data may be stored for future reuse.