公开(公告)号：US10454962B2

公开(公告)日：2019-10-22

申请号：US16159481

申请日：2018-10-12

Applicant: Apple Inc.

Inventor： Abhradeep Guha Thakurta ,  Andrew H. Vyrros ,  Umesh S. Vaishampayan ,  Gaurav Kapoor ,  Julien Freudiger ,  Vipul Ved Prakash ,  Arnaud Legendre ,  Steven Duplinsky

IPC: H01L29/06 ,  G06N20/00 ,  G06F17/27 ,  G06F17/22 ,  G06F21/62 ,  H04L29/06

Abstract: Systems and methods are disclosed for generating term frequencies of known terms based on crowdsourced differentially private sketches of the known terms. An asset catalog can be updated with new frequency counts for known terms based on the crowdsourced differentially private sketches. Known terms can have a classification. A client device can maintain a privacy budget for each classification of known terms. Classifications can include emojis, deep links, locations, finance terms, and health terms, etc. A privacy budget ensures that a client does not transmit too much information to a term frequency server, thereby compromising the privacy of the client device.

公开(公告)号：US10229282B2

公开(公告)日：2019-03-12

申请号：US15275284

申请日：2016-09-23

Applicant: Apple Inc.

Inventor： Yannick L. Sierra ,  Abhradeep Guha Thakurta ,  Umesh S. Vaishampayan ,  John C. Hurley ,  Keaton F. Mowery ,  Michael Brouwer

IPC: G06F21/62 ,  H04L9/06 ,  H04L29/06 ,  H04L9/08

Abstract: The system described may implement a 1-bit protocol for differential privacy for a set of client devices that transmit information to a server. Implementations of the system may leverage specialized instruction sets or engines built into the hardware or firmware of a client device to improve the efficiency of the protocol. For example, a client device may utilize these cryptographic functions to randomize information sent to the server. In one embodiment, the client device may use cryptographic functions such as hashes including SHA or block ciphers including AES. Accordingly, the system provides an efficient mechanism for implementing differential privacy.

公开(公告)号：US10133725B2

公开(公告)日：2018-11-20

申请号：US15477921

申请日：2017-04-03

Applicant: Apple Inc.

Inventor： Abhradeep Guha Thakurta ,  Andrew H. Vyrros ,  Umesh S. Vaishampayan ,  Gaurav Kapoor ,  Julien Freudiger ,  Vivek Rangarajan Sridhar ,  Doug Davidson

IPC: G06F17/27 ,  G06N99/00 ,  G06F17/16 ,  G06F17/30

Abstract: Systems and methods are disclosed for a server learning new words generated by user client devices in a crowdsourced manner while maintaining local differential privacy of client devices. A client device can determine that a word typed on the client device is a new word that is not contained in a dictionary or asset catalog on the client device. New words can be grouped in classifications such as entertainment, health, finance, etc. A differential privacy system on the client device can comprise a privacy budget for each classification of new words. If there is privacy budget available for the classification, then one or more new terms in a classification can be sent to new term learning server, and the privacy budget for the classification reduced. The privacy budget can be periodically replenished.

公开(公告)号：US09942355B2

公开(公告)日：2018-04-10

申请号：US15152776

申请日：2016-05-12

Applicant: Apple Inc.

Inventor： Michael R. Siracusa ,  Joao Pedro Forjaz de Lacerda ,  Kit-Man Wan ,  Gaurav Kapoor ,  Umesh S. Vaishampayan

IPC: G06F9/44 ,  H04L29/08 ,  H04W24/02 ,  G06N7/00 ,  G06F9/445 ,  G06F9/54 ,  G06F11/30 ,  H04W52/02

CPC classification number: H04L67/327 ,  G06F8/61 ,  G06F9/445 ,  G06F9/542 ,  G06F9/546 ,  G06F11/30 ,  G06N7/005 ,  H04L67/14 ,  H04L67/26 ,  H04W24/02 ,  H04W52/0264 ,  Y02D10/43 ,  Y02D70/00 ,  Y02D70/142 ,  Y02D70/144 ,  Y02D70/164 ,  Y02D70/26

Abstract: Disclosed are systems, methods, and non-transitory computer-readable storage media for monitoring the current context of a computing device. In some implementations, a context daemon can collect context information about the computing device. The context information can include current device hardware state information. The context information can include current software state information. The context can be derived or implied from a combination of hardware state information, software state information, or any other type of state information. For example, the derived context can be a user state (e.g., a user activity, sleeping, running, etc.) derived from or implied by hardware or software state information.

公开(公告)号：US20180039619A1

公开(公告)日：2018-02-08

申请号：US15477921

申请日：2017-04-03

Applicant: Apple Inc.

Inventor： Abhradeep Guha Thakurta ,  Andrew H. Vyrros ,  Umesh S. Vaishampayan ,  Gaurav Kapoor ,  Julien Freudiger ,  Vivek Rangarajan Sridhar ,  Doug Davidson

IPC: G06F17/27 ,  G06F17/16 ,  G06N99/00

CPC classification number: G06F17/2765 ,  G06F17/16 ,  G06F17/2705 ,  G06F17/2735 ,  G06F17/277 ,  G06F17/30737 ,  G06N99/005

Abstract: Systems and methods are disclosed for a server learning new words generated by user client devices in a crowdsourced manner while maintaining local differential privacy of client devices. A client device can determine that a word typed on the client device is a new word that is not contained in a dictionary or asset catalog on the client device. New words can be grouped in classifications such as entertainment, health, finance, etc. A differential privacy system on the client device can comprise a privacy budget for each classification of new words. If there is privacy budget available for the classification, then one or more new terms in a classification can be sent to new term learning server, and the privacy budget for the classification reduced. The privacy budget can be periodically replenished.

公开(公告)号：US20170357820A1

公开(公告)日：2017-12-14

申请号：US15275284

申请日：2016-09-23

Applicant: Apple Inc.

Inventor： Yannick L. Sierra ,  Abhradeep Guha Thakurta ,  Umesh S. Vaishampayan ,  John C. Hurley ,  Keaton F. Mowery ,  Michael Brower

IPC: G06F21/62 ,  H04L29/06 ,  H04L9/06

CPC classification number: G06F21/6218 ,  G06F21/6245 ,  H04L9/0631 ,  H04L9/0861 ,  H04L63/0421 ,  H04L63/0435

Abstract: The system described may implement a 1-bit protocol for differential privacy for a set of client devices that transmit information to a server. Implementations of the system may leverage specialized instruction sets or engines built into the hardware or firmware of a client device to improve the efficiency of the protocol. For example, a client device may utilize these cryptographic functions to randomize information sent to the server. In one embodiment, the client device may use cryptographic functions such as hashes including SHA or block ciphers including AES. Accordingly, the system provides an efficient mechanism for implementing differential privacy.

公开(公告)号：US09465679B2

公开(公告)日：2016-10-11

申请号：US14622674

申请日：2015-02-13

Applicant: Apple Inc.

Inventor： Phillip Stanley-Marbell ,  Gaurav Kapoor ,  Umesh S. Vaishampayan

IPC: G06F9/54

CPC classification number: G06F9/542 ,  G06F9/4843 ,  G06F9/4893 ,  G06F2209/482 ,  Y02D10/24

Abstract: In some implementations, a mobile device can be configured to monitor environmental, system and user events associated with the mobile device and/or a peer device. The occurrence of one or more events can trigger adjustments to system settings. The mobile device can be configured to keep frequently invoked applications up to date based on a forecast of predicted invocations by the user. In some implementations, the mobile device can receive push notifications associated with applications that indicate that new content is available for the applications to download. The mobile device can launch the applications associated with the push notifications in the background and download the new content. In some implementations, before running an application or communicating with a peer device, the mobile device can be configured to check energy and data budgets and environmental conditions of the mobile device and/or a peer device to ensure a high quality user experience.

Abstract translation: 在一些实现中，移动设备可以被配置为监视与移动设备和/或对等设备相关联的环境，系统和用户事件。 一个或多个事件的发生可以触发对系统设置的调整。 移动设备可以被配置为基于用户预测的调用的预测来保持频繁被调用的应用程序的最新。 在一些实现中，移动设备可以接收与指示新内容可用于应用下载的应用相关联的推送通知。 移动设备可以在后台启动与推送通知相关联的应用，并下载新的内容。 在一些实施方案中，在运行应用程序或与对等设备通信之前，移动设备可以被配置为检查移动设备和/或对等设备的能量和数据预算以及环境状况，以确保高质量的用户体验。

公开(公告)号：US12020168B2

公开(公告)日：2024-06-25

申请号：US16262807

申请日：2019-01-30

Applicant: Apple Inc.

Inventor： Francesco Rossi ,  Cecile M. Foret ,  Gaurav Kapoor ,  Kit-Man Wan ,  Umesh S. Vaishampayan ,  Etienne Belanger

IPC: G06N3/10 ,  G06F9/46 ,  G06F9/48 ,  G06F9/50

CPC classification number: G06N3/10 ,  G06F9/461 ,  G06F9/4881 ,  G06F9/5038

Abstract: The subject technology runs a compiled neural network (NN) model on a particular processor with multiple priority queues for executing different processes, the compiled NN model being assigned to a particular priority queue, and the compiled NN model includes context switch instructions that were previously inserted into a neural network (NN) model from which the compiled NN model was compiled. The subject technology determines that a particular context switch instruction has been executed by the particular processor. The subject technology determines that a different process is waiting to be executed, the different process being assigned to a different priority queue and the different process being a higher priority process than the running compiled NN model. In response to executing the particular context switch instruction, the subject technology performs a context switch to the different process assigned to the different priority queue when the different process is waiting to be executed.

公开(公告)号：US11710035B2

公开(公告)日：2023-07-25

申请号：US16556066

申请日：2019-08-29

Applicant: Apple Inc.

Inventor： Abhishek Bhowmick ,  Ryan M. Rogers ,  Umesh S. Vaishampayan ,  Andrew H. Vyrros

IPC: G06N3/08 ,  G06N3/04 ,  G06N3/10 ,  G06N3/044 ,  G06N3/045

CPC classification number: G06N3/08 ,  G06N3/04 ,  G06N3/044 ,  G06N3/045 ,  G06N3/10

Abstract: Embodiments described herein provide a technique to crowdsource labeling of training data for a machine learning model while maintaining the privacy of the data provided by crowdsourcing participants. Client devices can be used to generate proposed labels for a unit of data to be used in a training dataset. One or more privacy mechanisms are used to protect user data when transmitting the data to a server. The server can aggregate the proposed labels and use the most frequently proposed labels for an element as the label for the element when generating training data for the machine learning model. The machine learning model is then trained using the crowdsourced labels to improve the accuracy of the model.

公开(公告)号：US11042664B2

公开(公告)日：2021-06-22

申请号：US16746683

申请日：2020-01-17

Applicant: Apple Inc.

Inventor： Yannick L. Sierra ,  Abhradeep Guha Thakurta ,  Umesh S. Vaishampayan ,  John C. Hurley ,  Keaton F. Mowery ,  Michael Brouwer

IPC: G06F21/62 ,  H04L9/06 ,  H04L9/08 ,  H04L29/06

Abstract: One embodiment provides a system that implements a 1-bit protocol for differential privacy for a set of client devices that transmit information to a server. Implementations may leverage specialized instruction sets or engines built into the hardware or firmware of a client device to improve the efficiency of the protocol. For example, a client device may utilize these cryptographic functions to randomize information sent to the server. In one embodiment, the client device may use cryptographic functions such as hashes including SHA or block ciphers including AES to provide an efficient mechanism for implementing differential privacy.