INSERTING SECONDARY CONTENT INTO PRIMARY CONTENT PRESENTED BY AN ONLINE SYSTEM
Granted: January 14, 2021
Application Number:
20210014544
An online system receives a request for a video to be presented by the online system to a target user. The online system determines whether to insert secondary content into the video. For such a determination, the online system identifies a position in the video for inserting secondary content. Further, the online system determines a loss score and a gain score. The loss score measures a loss of interaction by the target user if the secondary content were inserted. The gain score…
SYSTEMS AND METHODS FOR ASYMMETRICAL SCALING FACTOR SUPPORT FOR NEGATIVE AND POSITIVE VALUES
Granted: January 14, 2021
Application Number:
20210012202
Disclosed herein includes a system, a method, and a device for asymmetrical scaling factor support for negative and positive values. A device can include a circuit having a shift circuitry and multiply circuitry. The circuit can be configured to perform computation for a neural network, including multiplying, via the multiply circuitry, a first value and a second value. The circuit can be configured to perform computation for a neural network, including shifting, via the shift circuitry,…
SYSTEMS AND METHODS FOR PIPELINED PARALLELISM TO ACCELERATE DISTRIBUTED PROCESSING
Granted: January 14, 2021
Application Number:
20210012186
Disclosed herein includes a system, a method, and a device for pipelined parallelism to accelerate distributed learning network graph. First data for a first layer of a neural network may be stored in memory. First circuitry including a first plurality of processing element (PE) circuits may read the first data from the memory and perform computation for the first layer of the neural network using the first data to generate second data. The first circuitry includes a plurality of buffers…
SYSTEMS, METHODS, AND DEVICES FOR EARLY-EXIT FROM CONVOLUTION
Granted: January 14, 2021
Application Number:
20210012178
Disclosed herein includes a system, a method, and a device for early-exit from convolution. In some embodiments, at least one processing element (PE) circuit is configured to perform, for a node of a neural network corresponding to a dot-product operation with a set of operands, computation using a subset of the set of operands to generate a dot-product value of the subset of the set of operands. The at least one PE circuit can compare the dot-product value of the subset of the set of…
POWER EFFICIENT MULTIPLY-ACCUMULATE CIRCUITRY
Granted: January 14, 2021
Application Number:
20210011971
Disclosed herein includes a system, a method, and a device for multiply-accumulate operation. In one aspect, an input operand is received by control circuitry. In one aspect, the control circuitry determines a sparsity of the input operand, where the sparsity may indicate whether a value of the input operand has a predetermined value or not. In one aspect, the control circuitry determines a stationarity of the input operand, where the stationarity may indicate whether the value of the…
SYSTEMS AND METHODS FOR READING AND WRITING SPARSE DATA IN A NEURAL NETWORK ACCELERATOR
Granted: January 14, 2021
Application Number:
20210011846
Disclosed herein includes a system, a method, and a device for reading and writing sparse data in a neural network accelerator. A plurality of slices can be established to access a memory having an access size of a data word. A first slice can be configured to access a first side of the data word in memory. Circuitry can access a mask identifying byte positions within the data word having non-zero values. The circuitry can modify the data word to have non-zero byte values stored starting…
SYSTEMS AND METHODS FOR DISTRIBUTING A NEURAL NETWORK ACROSS MULTIPLE COMPUTING DEVICES
Granted: January 14, 2021
Application Number:
20210011288
Disclosed herein is a method for using a neural network across multiple devices. The method can include receiving, by a first device configured with a first one or more layers of a neural network, input data for processing via the neural network implemented across the first device and a second device. The method can include outputting, by the first one or more layers of the neural network implemented on the first device, a data set that is reduced in size relative to the input data while…
INFORMATIVE ADVERTISEMENTS ON HOBBY AND STRONG INTERESTS FEATURE SPACE
Granted: January 7, 2021
Application Number:
20210004872
Information for an advertising campaign is received that includes one or more advertisements for presentation to one or more users of an online system. A strong interest feature domain is generated for each user of the online system. A set of related items is obtained, wherein each item in the set is associated with an action conducted by the one or more users. It can be determined that a viewing user of the online system has interacted with at least one item in the set of related items,…
COUNTER BASED MULTIPLY-AND-ACCUMULATE CIRCUIT FOR NEURAL NETWORK
Granted: January 7, 2021
Application Number:
20210004208
Disclosed herein includes a system, a method, and a device for improving computation efficiency of a neural network. In one aspect, adder circuitry is configured to add input data from processing of the neural network and a first number of bits of accumulated data for the neural network to generate summation data. In one aspect, according to a carry value of the adding from the adder circuitry, a multiplexer is configured to select between i) a second number of bits of the accumulated…
REAL-TIME TRACKING OF OFFLINE TRANSACTIONS
Granted: December 17, 2020
Application Number:
20200394681
An online system receives offline conversion data in real-time from a third party system regarding an offline user, the offline conversion data indicating an action performed by the offline user and identifying information for the offline user. The online system identifies a local user matching the offline user based on the identifying information for the offline user. The online system stores the offline conversion for the identified local user. The online system determines an…
AUTOMATED VISUAL SUGGESTION, GENERATION, AND ASSESSMENT USING COMPUTER VISION DETECTION
Granted: October 15, 2020
Application Number:
20200327375
An online system may identify content with which a user has an interest. For example, the online system may determine that a user has an interest in the content based on interaction information indicating that the user interacted with the content. In a particular example, the online system may identify image concepts included in the content based on computer vision techniques that recognize the image concepts. The online system may model probabilities that image concepts will appeal to…
Grouping Content Based on Geographic Data
Granted: October 1, 2020
Application Number:
20200314166
A method includes detecting content items stored locally on a client device, each content item including a time stamp and geospatial metadata; identifying one or more sets of content items based on their time stamps being within a first predetermined time range; determining, for each set of content items, a geographic location associated with the content items of the set based on the geospatial metadata of each content item; generating a list of all geographic locations associated with…
SENTIMENT POLARITY FOR USERS OF A SOCIAL NETWORKING SYSTEM
Granted: September 10, 2020
Application Number:
20200286000
A social networking system infers a sentiment polarity of a user toward content of a page. The sentiment polarity of the user is inferred based on received information about an interaction between the user and the page (e.g., like, report, etc.), and may be based on analysis of a topic extracted from text on the page. The system infers a positive or negative sentiment polarity of the user toward the content of the page, and that sentiment polarity then may be associated with any second…
DIALER WITH REAL-TIME REVERSE LOOK-UP INCLUDING SOCIAL DATA
Granted: September 3, 2020
Application Number:
20200280629
Methods and systems that include, by a client computing device of a first user of a social-networking system, receiving from the first user an identifier associated with an intended recipient to which outgoing communication is to be initiated, wherein the identifier is received via a user interface of an application, identifying, based on the intended recipient, social information relevant to the first user, and displaying, in the user interface, the social information that is relevant…
METHODS AND APPARATUS FOR UNSUPERVISED ONE-SHOT MACHINE LEARNING FOR CLASSIFICATION OF HUMAN GESTURES AND ESTIMATION OF APPLIED FORCES
Granted: September 3, 2020
Application Number:
20200275895
Methods and apparatus for training a classification model and using the trained classification model to recognize gestures performed by a user. An apparatus comprises a processor that is programmed to: receive, via a plurality of neuromuscular sensors, a first plurality of neuromuscular signals from a user as the user performs a first single act of a gesture; train a classification model based on the first plurality of neuromuscular signals, the training including: deriving value(s) from…
EVALUATING MODIFICATIONS TO FEATURES USED BY MACHINE LEARNED MODELS APPLIED BY AN ONLINE SYSTEM
Granted: August 27, 2020
Application Number:
20200272943
An online system identifies an additional feature to evaluate for inclusion in a machine learned model. The additional feature is based on characteristics of one or more dimensions of information maintained by the online system. To generate data for evaluating the additional feature, the online system generates various partitions of stored data, where each partition includes characteristics associated with one or more dimensions on which the additional feature is based. Using values of…
CONVERSION OPTIMIZATION WITH LONG ATTRIBUTION WINDOW
Granted: August 20, 2020
Application Number:
20200265471
An online system optimizes for longer attribution window conversions with an additive decomposition model by predicting the probability that a predefined action happens given an impression/click. The online system receives a content item from a content provider for display to a target user, and predicts a probability that a target user will convert given an interaction with the content item by the target user. The online system computes, by a first trained model, a short-term conversion…
CLUSTERING USERS OF A SOCIAL NETWORKING SYSTEM BASED ON USER INTERACTIONS WITH CONTENT ITEMS ASSOCIATED WITH A TOPIC
Granted: August 6, 2020
Application Number:
20200250702
A social networking system presents users with a content items and ad requests, which may include targeting criteria specifying a topic. Interactions by users who were presented with an advertisement from an ad request including targeting criteria specifying the topic are stored by the social networking system and used to identify a cluster group of additional users having characteristics similar to characteristics of users who were presented with the advertisement from the ad request…
Re-Ranking Story Content
Granted: August 6, 2020
Application Number:
20200249784
In one embodiment, a method includes receiving a cover feed interaction history from a device associated with a user of a social-networking system. An order for a plurality of content boards may be determined based on the cover feed interaction history (e.g., viewing history, download status, the current order of content boards in the cover feed, user interaction history, whether the user bookmarked or pinned a particular content board), user information related to the user, and device…
MESSENGER MSQRD - MASK INDEXING
Granted: July 16, 2020
Application Number:
20200227090
Exemplary embodiments relate to the creation of a media effect index for group video conversations. Media effect application (e.g., in the form of graphical overlays, filters, sounds, etc.) may be tracked in a timeline during a chat session. The resulting index may be used to create a highlights reel, which may serve as an index into a live show or may be used to determine the best time to insert materials into a recording of the conversation. The index may be used to automatically…
