Abhishek Sharma

8619464, Dec 31, 2013

Aug 26, 2011

13/219537

Shankar Sinha - Redwood City CA, US
Brian Wong - Sunnyvale CA, US
Shih-Lin S. Lee - San Jose CA, US
Abhishek Sharma - San Jose CA, US

Altera Corporation - San Jose CA

G11C 11/00

365154, 365156, 36518914

Integrated circuits may have arrays of memory elements. Data may be loaded into the memory elements and read from the memory elements using data lines. Address lines may be used to apply address signals to write address transistors and read circuitry. A memory element may include a bistable storage element. Read circuitry may be coupled between the bistable storage element and a data line. The read circuitry may include a data storage node. A capacitor may be coupled between the data storage node and ground and may be used in storing preloaded data from the bistable storage element. The read circuitry may include a transistor that is coupled between the bistable storage element and the data storage node and a transistor that is coupled between the data storage node and the data line.

20220222441, Jul 14, 2022

Mar 15, 2021

17/202183

- San Francisco CA, US
Abhishek Sharma - Fremont CA, US
Suhail Sanjiv Barot - San Mateo CA, US
Gurkirat Singh - Elk Grove CA, US
Mridul Gupta - San Mateo CA, US
Shiva Kumar Pentyala - Austin TX, US
Ankit Chadha - San Jose CA, US

G06F 40/295
G06F 40/35
G06F 40/247
G06N 3/08

A system performs named entity recognition for performing natural language processing, for example, for conversation engines. The system uses context information in named entity recognition. The system includes the context of a sentence during model training and execution. The system generates high quality contextual data for training NER models. The system utilizes labeled and unlabeled contextual data for training NER models. The system provides NER models for execution in production environments. The system uses heuristics to determine whether to use a context-based NER model or a simple NER model that does not use context information. This allows the system to use simple NER models when the likelihood of improving the accuracy of prediction based on context is low.

20220222489, Jul 14, 2022

Mar 15, 2021

17/202188

- San Francisco CA, US
Abhishek Sharma - Fremont CA, US
Suhail Sanjiv Barot - San Mateo CA, US
Gurkirat Singh - Elk Grove CA, US
Mridul Gupta - San Mateo CA, US
Shiva Kumar Pentyala - Austin TX, US
Ankit Chadha - San Jose CA, US

G06K 9/62
G06F 40/295
G06F 40/247
G06F 40/35
G06F 40/284
G06N 20/00

A system performs named entity recognition for performing natural language processing, for example, for conversation engines. The system uses context information in named entity recognition. The system includes the context of a sentence during model training and execution. The system generates high quality contextual data for training NER models. The system utilizes labeled and unlabeled contextual data for training NER models. The system provides NER models for execution in production environments. The system uses heuristics to determine whether to use a context-based NER model or a simple NER model that does not use context information. This allows the system to use simple NER models when the likelihood of improving the accuracy of prediction based on context is low.

20230086302, Mar 23, 2023

Sep 20, 2021

17/479748

- San Francisco CA, US
Shubham Mehrotra - Santa Clara CA, US
Abhishek Sharma - Fremont CA, US
Shashank Harinath - Santa Clara CA, US
Na Cheng - Yarrow Point WA, US
Zineb Laraki - Atherton CA, US

G06F 40/263
G06F 40/58
G06K 9/62

A method that includes receiving an input at an interactive conversation service that uses an intent classification model. The method may further include generating, using an encoder model of the intent classification model, a set of output vectors corresponding to the input, where the encoder model is configured to determine a set of metrics corresponding to intent classifications. The method may further include determining, using an outlier detection model of the intent classification model, whether the input is in-domain or out-of-domain (OOD) based on a first vector of the set of output vectors satisfying a domain threshold relative to one or more of the intent classifications. The method may further include outputting, by the intent classification model, a second vector of the set of output vectors that indicates the set of metrics corresponding to the intent classifications or an indication that the input is OOD.

20210166691, Jun 3, 2021

Dec 23, 2020

17/132439

- Cupertino CA, US
Abhishek Sharma - Sunnyvale CA, US
Jerremy Holland - Sunnyvale CA, US
Nicholas E. Apostoloff - San Jose CA, US
Russell Y. Webb - San Jose CA, US
Tomas J. Pfister - Foster City CA, US

G10L 15/22
G06K 9/00
G10L 15/18
G10L 15/26

Modifying operation of an intelligent agent in response to facial expressions and/or emotions.

20200357449, Nov 12, 2020

Sep 28, 2017

16/629466

- Santa Clara CA, US
Prashant MAJHI - San Jose CA, US
Kaan OGUZ - Beaverton OR, US
Kevin P. O'BRIEN - Portland OR, US
Abhishek A. SHARMA - Hillsboro OR, US
David L. KENCKE - Beaverton OR, US

G11C 11/16
H01L 43/08
H01L 43/02
H01L 43/10
H01L 45/00

Spin transfer torque memory (STTM) devices incorporating an Insulator-Metal-Transition (IMT) device or at least one layer of Insulator-Metal-Transition (IMT) material are disclosed. The Insulator-Metal-Transition (IMT) device or at least one layer of Insulator-Metal-Transition (IMT) material are utilized for providing a spike current when the voltage across it exceeds the threshold voltage to reduce a critical current required for transfer torque induced magnetization switching.

20200220023, Jul 9, 2020

Jan 4, 2018

16/649171

- Santa Clara CA, US
Prashant Majhi - San Jose CA, US
Seung Hoon Sung - Portland OR, US
Willy Rachmady - Beaverton OR, US
Gilbert Dewey - Hillsboro OR, US
Abhishek A. Sharma - Hillsboro OR, US
Brian S. Doyle - Portland OR, US
Jack T. Kavalieros - Portland OR, US

H01L 29/786
H01L 29/66
H01L 29/417
H01L 29/36
H01L 25/065
H01L 21/225
H01L 27/24

An embodiment includes a system comprising: a thin film transistor (TFT) comprising a source, a channel, a drain, and a gate; first, second, and third dielectric portions; wherein (a) a first vertical axis intersects the source, the channel, and the drain; (b) the first dielectric portion surrounds the source in a first plane; (c) the second dielectric portion surrounds the channel in a second plane; (d) the third dielectric surrounds the drain in a third plane; (e) a second vertical axis intersects the first, second, and third dielectric portions; (f) the source includes a first dopant, the first dielectric portion includes the first dopant, the second dielectric portion includes at least one of the first dopant and a second dopant, the drain includes the at least one of the first and second dopants, and the third dielectric portion includes the at least one of the first and second dopants.

20200192640, Jun 18, 2020

Feb 25, 2020

16/801099

- Sunnyvale CA, US
Abhishek Sharma - Mountain View CA, US

G06F 8/35
G06F 9/48
G06F 9/448
G06F 16/9535
G06F 11/36
G06F 8/40
G06F 8/34
G06F 8/30

An efficient state-machine-based pattern matching technique processes tokens in an input queue and identifies patterns in the sequence of tokens that match one or more predetermined input patterns without backtracking. Tokens can include data or no data and a time component. The tokens can be a stream of data generated by a sensor, which transforms a physical property into a digital quantity. The pattern matching technique processes the input queue in a single direction, and does not examine any previously examined token. In an implementation, specific patterns to be matched are specified using a state machine, where the state machine is specified in a state table and operates using a state stack.