Sheng Chieh Yang

US Patent:

20220217164, Jul 7, 2022

Filed:

Mar 23, 2022

Appl. No.:

17/702687

Inventors:

- Santa Clara CA, US
Suiqiang Deng - Fremont CA, US
Sheng Yang - Santa Clara CA, US
Ho Yu Lam - Santa Clara CA, US

International Classification:

H04L 9/40
G06N 5/02

Abstract:

Detection of malicious files is disclosed. A set comprising a plurality of sample classification models is received and stored. A determination is made that n-gram analysis should be performed on a sequence of received packets associated with a received file. Performing the n-gram analysis includes using a determined filetype associated with the sequence of received packets to select at least one stored sample classification model included in the set for use in performing the n-gram analysis. A determination is made that the received file is malicious based at least in part on the n-gram analysis of the sequence of received packets. In response to determining that the file is malicious, propagation of the received file is prevented.

US Patent:

20210124350, Apr 29, 2021

Filed:

Oct 23, 2019

Appl. No.:

16/661402

Inventors:

- San Francisco CA, US
David Tse-Zhou Lu - Menlo Park CA, US
Sheng Yang - Palo Alto CA, US
Ranjith Unnikrishnan - Fremont CA, US
Emilie Jeanne Anne Danna - Sunnyvale CA, US
Weiyi Hou - Mountain View CA, US
Suneet Rajendra Shah - San Francisco CA, US
Ying Liu - Los Altos CA, US

Assignee:

Lyft, Inc. - San Francisco CA

International Classification:

G05D 1/00

Abstract:

Systems, methods, and non-transitory computer-readable media can access a plurality of parameter-based encodings providing a structured representation of an environment captured by one or more sensors associated with a plurality of vehicles traveling through the environment. A given parameter-based encoding of the environment identifies one or more agents that were detected by a vehicle within the environment and respective location information for the one or more agents within the environment. The plurality of parameter-based encodings can be clustered into one or more clusters of parameter-based encodings. At least one scenario associated with the environment can be determined based at least in part on the one or more clusters of parameter-based encodings.

US Patent:

20210124355, Apr 29, 2021

Filed:

Oct 23, 2019

Appl. No.:

16/661373

Inventors:

- San Francisco CA, US
David Tse-Zhou Lu - Menlo Park CA, US
Sheng Yang - Palo Alto CA, US
Ranjith Unnikrishnan - Fremont CA, US
Emilie Jeanne Anne Danna - Sunnyvale CA, US
Weiyi Hou - Mountain View CA, US
Suneet Rajendra Shah - San Francisco CA, US
Ying Liu - Los Altos CA, US

Assignee:

Lyft, Inc. - San Francisco CA

International Classification:

G05D 1/02
G05D 1/00

Abstract:

Systems, methods, and non-transitory computer-readable media can determine sensor data captured by at least one sensor of a vehicle while navigating an environment over a period of time. Information describing one or more agents associated with the environment during the period of time can be determined based at least in part on the captured sensor data. A parameter-based encoding describing the one or more agents associated with the environment during the period of time can be generated based at least in part on the determined information and a scenario schema, wherein the parameter-based encoding provides a structured representation of the information describing the one or more agents associated with the environment. A scenario represented by the parameter-based encoding can be determined based at least in part on a cluster of parameter-based encodings to which the parameter-based encoding is assigned.

US Patent:

20210063175, Mar 4, 2021

Filed:

Sep 23, 2020

Appl. No.:

16/948576

Inventors:

- San Francisco CA, US
Upamanyu Madhow - Santa Barbara CA, US
Vikram Saxena - Cupertino CA, US
Livia Zarnescu Yanez - Menlo Park CA, US
Chandan Prakash Sheth - Fremont CA, US
Sheng Yang - Fremont CA, US
Alvin AuYoung - San Jose CA, US

International Classification:

G01C 21/34
G01C 21/36
G01C 21/20
G06N 5/02
G06F 16/29

Abstract:

Systems and methods of using sensor data for coordinate prediction are disclosed herein. In some example embodiments, for a place, a computer system accesses corresponding service data comprising pick-up data and drop-off data for requests, and accesses corresponding sensor data indicating at least one path of mobile devices of the requesters of the requests, with the at least one path comprising at least one of a pick-up path ending at the pick-up location indicated by the pick-up data and a drop-off path beginning at the drop-off location indicated by the drop-off data. In some example embodiments, the computer system generates at least one predicted geographic location using the paths indicated by the sensor data, and stores the at least one predicted geographic location in a database in association with an identification of the place.

US Patent:

20210019412, Jan 21, 2021

Filed:

Jul 19, 2019

Appl. No.:

16/517465

Inventors:

- Santa Clara CA, US
Suiqiang Deng - Fremont CA, US
Sheng Yang - Santa Clara CA, US
Ho Yu Lam - Santa Clara CA, US

International Classification:

G06F 21/56
G06N 20/00

Abstract:

Generating models usable by data appliances to perform inline malware analysis is disclosed. A set of features, including a plurality of n-grams, extracted from a set of files is received. A reduced set of features is determined that includes at least some of the plurality of n-grams. The reduced set of features is used to generate a model usable by a data appliance to perform inline malware analysis.

US Patent:

20210021611, Jan 21, 2021

Filed:

Jul 19, 2019

Appl. No.:

16/517463

Inventors:

- Santa Clara CA, US
Suiqiang Deng - Fremont CA, US
Sheng Yang - Santa Clara CA, US
Ho Yu Lam - Santa Clara CA, US

International Classification:

H04L 29/06
G06N 5/02

Abstract:

Detection of malicious files is disclosed. A set comprising one or more sample classification models is stored on a networked device. N-gram analysis is performed on a sequence of received packets associated with a received file. Performing the n-gram analysis includes using at least one stored sample classification model. A determination is made that the received file is malicious based at least in part on the n-gram analysis of the sequence of received packets. In response to determining that the file is malicious, propagation of the received file is prevented.

US Patent:

20200080852, Mar 12, 2020

Filed:

Sep 6, 2018

Appl. No.:

16/123656

Inventors:

- San Francisco CA, US
Sheng Yang - Fremont CA, US

International Classification:

G01C 21/34
G06N 3/08
G06N 3/04
G06K 9/62
G06T 7/00
G01C 21/20

Abstract:

Systems and methods of identifying incorrect coordinate prediction using route information are disclosed herein. In some example embodiments, a computer system receives route information and geographic image data. The route information corresponds to a servicing of a request associated with a place and indicates a travelled route that has been traveled by a user in traveling from an origin location to the place, and the geographic image data represents a geographic area corresponding to the travelled route. The computer system determines that an initial geographic location stored in a database in association with the place is incorrect based on the route information and the geographic image data using a first deep learning model, and then performs a verification operation based on the determining that the stored initial geographic location of the place is incorrect.

US Patent:

20200041298, Feb 6, 2020

Filed:

Oct 14, 2019

Appl. No.:

16/601510

Inventors:

- San Francisco CA, US
Sheng Yang - Fremont CA, US
Vasudev Parameswaran - Fremont CA, US
Shivendra Pratap Singh - Redwood CIty CA, US
Jane Alam Jan - Santa Clara CA, US

International Classification:

G01C 21/34
G06F 7/14
G06F 7/02
G06F 7/20
G01C 21/36

Abstract:

An authoritative candidate is selected for determining a location of a point of interest (POI). Source data including name, address, and location for POIs is received from multiple data sources. The received data is normalized for ease of comparison, and coordinates for each candidate are compared to coordinates of other candidates to determine which candidate if any is an authoritative location for the POI. The candidate locations are compared using two models a metric-based scoring system and a machine learning model that may utilize a gradient boosted decision tree. The authoritative candidate can be used to render digital maps that include the POI. In addition, the authoritative candidate's location can be used to provide vehicle route guidance to the POI.