Abstract: Techniques are described herein for cold-start forecasting datasets using backcasting and composite embedding. An example method can include a system receiving a set of time series and metadata text comprising a first subset of metadata text and a second subset of metadata text. The system can generate a plurality of embeddings, each embedding comprising a numerical representation of a metadata text of the set of metadata text. The system can generate a plurality of vectors, each vector comprising a time series of the set of time series each time series associated with a metadata text of the first subset of metadata text. The system can generate a plurality of composite embeddings based at least in part on combining each embedding with a respective vector of the plurality of vectors. The system can determine a forecasted value associated with the second subset of metadata text based on the composite embeddings.

Abstract: Techniques described herein are directed toward univariate series truncation policy using change point detection. An example method can include a device determining a first time series comprising a first set of data points indexed over time. The device can determine a first and second change point of the first time series based on a relative position and a category of the change points. The device can generate a first and second truncated time series based on the change points. The device can generate a first and second forecasted value using a first forecasting technique. The device can compare the first forecasted value and the second forecasted value using a second time series. The device can select one of the forecasting techniques to generate a final forecasted value based on the comparison. The device can generate, using the selected first forecasting technique, the final forecasted value.

Abstract: Techniques are described herein for forecasting datasets using blend of temporal aggregation and grouped aggregation. An example method can include a device accessing a first and second time series, comprising a first data point associated with a first time step and a first value and a second data point associated with a second time step and a second value. The method can further include the device determining a grouped aggregated data point using the first and second time series by aligning the first and second data point. The method can further include the device determining the grouped aggregated data point by summing the first and second value. The method can further include determining a grouped aggregated time series. The method can further include the device determining a first set of input values for a machine learning model. The method can further include the device determining a first forecasted future value.

Abstract: A method for authenticating a transaction between two entities includes receiving, by one or more processors, a request associated with the transaction between a first entity associated with a first computing device and a second entity, generating, by the one or more processors, transaction details information associated with completion of the transaction between the first entity and the second entity, encrypting, by the one or more processors, the transaction details information by digitally signing the transaction details information, and transmitting, by the one or more processors, the transaction details information encrypted by the one or more processors, to the first computing device.

Abstract: A system for determining a calculation utilizing differential privacy including an interface and a processor. The interface is configured to receive a request to determine result data of a calculation using multitenanted data. The multitenanted data comprises tenant data associated with a plurality of tenants. The processor is configured to: determine the result data by performing the calculation on the multitenanted data; determine whether a deterministic modification is needed to ensure privacy based at least in part on whether a number of participants in the result data is less than a threshold; and in response to determining that the deterministic modification is needed to ensure privacy: determine the deterministic modification; numerically modify the result data using the deterministic modification to determine modified result data; and provide the modified result data.

Abstract: Provided are systems and methods for real-time identification of fraudulent users of an online resource such as a website or mobile application, including new user accounts that have yet to transact on the online resource. In one example, a method may include receiving, by a host platform of an online resource, a request from a user device associated with a user account of the online resource, creating, by the host platform, a device fingerprint of the user device based on a plurality of device attributes included in the request, determining, by the host platform, whether the device fingerprint matches a previously banned device fingerprint stored in a database by the online resource, and in response to a determination that the device fingerprint has been banned previously, automatically restricting, by the host platform, an ability of the user account with the online resource.

Abstract: Embodiments are disclosed for pixel-based techniques for combining vector graphics shapes. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving a selection of a plurality of vector graphics shapes to be combined based on an operation type, identifying a dominant shape based on the operation type, applying stroke and fill properties associated with the dominant shape to each of the plurality of vector graphics shapes, initializing a buffer at least as large as a bounding box associated with the plurality of vector graphics shapes, the buffer storing pixels inside the bounding box, and populating each pixel of the buffer with values representing pixel types determined from the plurality of vector graphics shapes based on the operation type.

Abstract: Techniques are described for performing inter-object interactions between vector objects to adjust the appearance of one vector object based on another vector object. For example, a vector object interaction framework may be implemented in which designers drag a vector object onto another vector object to trigger an interaction between the objects. Candidate interactions between pairs of object types may be pre-defined and stored in a suitable data structure. Thus, when one vector object is dragged onto a recipient vector object and a pause is detected, the object types for the dragged and recipient objects may be determined, and a corresponding set of candidate interactions for the pair of object types may be accessed, ranked, simulated, and/or presented as selectable previews. As such, one of the candidate interactions may be selected and executed to form one or more output vector objects that may be saved, exported, modified, and/or reused.

Abstract: Embodiments are disclosed for pixel-based techniques for combining vector graphics shapes. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving a selection of a plurality of vector graphics shapes to be combined based on an operation type, identifying a dominant shape based on the operation type, applying stroke and fill properties associated with the dominant shape to each of the plurality of vector graphics shapes, initializing a buffer at least as large as a bounding box associated with the plurality of vector graphics shapes, the buffer storing pixels inside the bounding box, and populating each pixel of the buffer with values representing pixel types determined from the plurality of vector graphics shapes based on the operation type.

Abstract: Techniques are described for performing inter-object interactions between vector objects to adjust the appearance of one vector object based on another vector object. For example, a vector object interaction framework may be implemented in which designers drag a vector object onto another vector object to trigger an interaction between the objects. Candidate interactions between pairs of object types may be pre-defined and stored in a suitable data structure. Thus, when one vector object is dragged onto a recipient vector object and a pause is detected, the object types for the dragged and recipient objects may be determined, and a corresponding set of candidate interactions for the pair of object types may be accessed, ranked, simulated, and/or presented as selectable previews. As such, one of the candidate interactions may be selected and executed to form one or more output vector objects that may be saved, exported, modified, and/or reused.

Abstract: A system for determining a calculation utilizing differential privacy including an interface and a processor. The interface is configured to receive a request to determine result data of a calculation using multitenanted data. The multitenanted data comprises tenant data associated with a plurality of tenants. The processor is configured to: determine the result data by performing the calculation on the multitenanted data; determine whether a deterministic modification is needed to ensure privacy based at least in part on whether a number of participants in the result data is less than a threshold; and in response to determining that the deterministic modification is needed to ensure privacy: determine the deterministic modification; numerically modify the result data using the deterministic modification to determine modified result data; and provide the modified result data.

Abstract: A system for determining a calculation utilizing differential privacy including an interface and a processor. The interface is configured to receive a request to determine a result of a calculation using multitenanted data. The processor is configured to determine result data by performing the calculation on the multitenanted data; determine a deterministic modification in the event that the deterministic modification is needed to ensure privacy; modify the result data using the deterministic modification to determine modified result data; and provide the modified result data.

Abstract: Vector art object deformation techniques applied to digital images by vector art rendering system of a computing device are described. The vector art rendering system is configured to deform vector art objects associated with anchor points within a digital image based solely on the movement of the anchor points. Moreover, the vector art rendering system is also configured to adjust anchor points associated with vector art objects within a digital image based solely on the movement of the vector art objects. Techniques are also described that facilitate the automatic association of path segment and vector art objects to anchor points that possess multiple preexisting vector art object associations.

Abstract: Vector art object deformation techniques applied to digital images by vector art rendering system of a computing device are described. The vector art rendering system is configured to deform vector art objects associated with anchor points within a digital image based solely on the movement of the anchor points. Moreover, the vector art rendering system is also configured to adjust anchor points associated with vector art objects within a digital image based solely on the movement of the vector art objects. Techniques are also described that facilitate the automatic association of path segment and vector art objects to anchor points that possess multiple preexisting vector art object associations.

Abstract: Systems and techniques for identifying and creating individual assets from a canvas containing an artwork include receiving a canvas containing an artwork. An organization of the artwork on the canvas is determined. Individual assets in the artwork on the canvas are identified by applying a segmentation rule based on the organization of the artwork. The individual assets are created.

Abstract: This disclosure covers systems and methods that sharpen the appearance of a digital illustration while moving the digital illustration. In certain embodiments, upon receiving a command to move a digital illustration, the disclosed systems and methods move the digital illustration (and its constituent line segments) to positions that both sharpen the appearance of the digital illustration and respond to the command. To facilitate sharpening the appearance of a digital illustration as part of a seemingly continuous movement, in some embodiments, the disclosed systems and methods move a blurry digital illustration to a position that sharpens the appearance of blurry axial-line segments by translating the digital illustration according to a translation vector and a command to move the digital illustration.

Abstract: This disclosure covers systems and methods that sharpen the appearance of a digital illustration while moving the digital illustration. In certain embodiments, upon receiving a command to move a digital illustration, the disclosed systems and methods move the digital illustration (and its constituent line segments) to positions that both sharpen the appearance of the digital illustration and respond to the command. To facilitate sharpening the appearance of a digital illustration as part of a seemingly continuous movement, in some embodiments, the disclosed systems and methods move a blurry digital illustration to a position that sharpens the appearance of blurry axial-line segments by translating the digital illustration according to a translation vector and a command to move the digital illustration.