Tajana Simunic Rosing

US Patent:

7246181, Jul 17, 2007

Filed:

Sep 14, 2004

Appl. No.:

10/941583

Inventors:

Tajana Simunic Rosing - Sunnyvale CA, US

Assignee:

Hewlett-Packard Development Company, L.P. - Houston TX

International Classification:

G06F 3/00
H04Q 7/20

US Classification:

710 18, 710 11, 4554522, 4555521

Abstract:

An electronic device includes multiple communication interfaces and a processor coupled to the interfaces. The processor is operable to identify an interface that can transfer data with a performance of a parameter that is closer to a desired performance level than the performance of the same parameter by another interface, and is operable to transfer the data via the identified interface. Such a device can, without operator input, select and transfer data via the communication interface that gives the best data-transfer performance relative to a particular parameter such as power consumption. Furthermore, if the performance of the selected communication interface changes during the data transfer, the device may, without operator input, identify an interface that offers better performance and switch over to transferring the data transfer via the identified interface. Moreover, the device may, without operator input, allow a number of software applications to share a lesser number of interfaces.

US Patent:

7272730, Sep 18, 2007

Filed:

Jul 31, 2003

Appl. No.:

10/632412

Inventors:

Andrea Acquaviva - Rovigo, IT
Luca Benini - Ferrara, IT
Tajana S. Rosing - Sunnyvale CA, US

Assignee:

Hewlett-Packard Development Company, L.P. - Houston TX

International Classification:

G06F 1/32

US Classification:

713300, 713320, 714 47

Abstract:

An improvement in a real time operating system for supporting at least one application, a processor and at least one hardware resource. The operating system includes a power manager layer. Such power manager layer is arranged to exchange information with an application, the processor and at least one hardware resource to provide real time power management responsive to the information.

US Patent:

20050170801, Aug 4, 2005

Filed:

Jan 30, 2004

Appl. No.:

10/769044

Inventors:

Vinay Deolalikar - Mountain View CA, US
Tajana Rosing - Sunnyvale CA, US

International Classification:

H04B017/00
H04B007/005
H04B007/01

US Classification:

455226100, 455067110, 455506000

Abstract:

A method controls the operation of devices which communicate over a wireless communications channel. The method includes determining a parameter of a received signal communicated over the wireless communications channel and determining a minimum threshold value of the received signal. An average duration of fade is determined using the parameter and the minimum threshold. The method detects whether the received signal is less than the minimum threshold value. At least one of the devices is placed in a sleep mode for approximately the average duration of fade in response to the received signal being detected as less than the minimum threshold value. The determined parameter of the received signal may be the root mean square value of the received signal.

US Patent:

20050171753, Aug 4, 2005

Filed:

Jan 30, 2004

Appl. No.:

10/769041

Inventors:

Tajana Rosing - Sunnyvale CA, US
Ozgur Celebican - Atlanta GA, US

International Classification:

G06F017/50

US Classification:

703018000

Abstract:

An arrangement and method provides energy consumption estimates of a multiple component electrical system controlled by a processor. The arrangement includes a component model corresponding to each component of the system. Each component model includes an energy consumption value for each one of a plurality of operating modes of its corresponding component. The operation of the system is then simulated on an operating cycle by operating cycle basis. A mode detector determines the operating mode of each system component during each cycle of the simulated operation and an energy consumption evaluator determines energy consumption of each component for each operating cycle. An accumulator then determines a total energy consumption of all of the system components.

US Patent:

20210334703, Oct 28, 2021

Filed:

Apr 21, 2021

Appl. No.:

17/236640

Inventors:

- Oakland CA, US
Mohsen Imani - San Diego CA, US
Behnam Khaleghi - San Diego CA, US
Tajana Rosing - San Diego CA, US

International Classification:

G06N 20/00
G06F 15/80

Abstract:

A method of defining an implementation of circuits in a programmable device can be provided by receiving a plurality of specifications for a hyperdimensional (HD) computing machine learning application for execution on a programmable device, determining parameters for a template architecture for HD computing machine learning using the plurality of specifications, the template architecture including an HD hypervector encoder, an HD associative search unit, programmable device pre-defined processing units, and programmable device pre-defined processing elements within the pre-defined processing units, and generating programmable device code configured to specify resources to be allocated within the programmable device using pre-defined circuits defined for use in the programmable device using the determined parameters for the template architecture.

US Patent:

20210326756, Oct 21, 2021

Filed:

Apr 7, 2021

Appl. No.:

17/224306

Inventors:

- Oakland CA, US
Tajana Simunic Rosing - San Diego CA, US
Mohsen Imani - San Diego CA, US
Sahand Salamat - San Diego CA, US

International Classification:

G06N 20/00
G06K 9/62

Abstract:

A method of providing a trained machine learning model can include providing a trained non-binary hyperdimensional machine learning model that includes a plurality of trained hypervector classes, wherein each of the trained hypervector classes includes N elements, and then, eliminating selected ones of the N elements from the trained non-binary hyperdimensional machine learning model based on whether the selected element has a similarity with other ones of the N elements, to provide a sparsified trained non-binary hyperdimensional machine learning model.

US Patent:

20200410404, Dec 31, 2020

Filed:

Jun 29, 2020

Appl. No.:

16/915643

Inventors:

- Oakland CA, US
Tajana Rosing - San Diego CA, US
Farinaz Koushanfar - La Jolla CA, US
Mohammad Sadegh Riazi - Los Angeles CA, US

International Classification:

G06N 20/20
H04L 9/08

Abstract:

A computing system can include a plurality of clients located outside a cloud-based computing environment, where each of the clients may be configured to encode respective original data with a respective unique secret key to generate data hypervectors that encode the original data. A collaborative machine learning system can operate in the cloud-based computing environment and can be operatively coupled to the plurality of clients, where the collaborative machine learning system can be configured to operate on the data hypervectors that encode the original data to train a machine learning model operated by the collaborative machine learning system or to generate an inference from the machine learning model.