Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media for generating tags for an object portrayed in a digital image based on predicted attributes of the object. For example, the disclosed systems can utilize interleaved neural network layers of alternating inception layers and dilated convolution layers to generate a localization feature vector. Based on the localization feature vector, the disclosed systems can generate attribute localization feature embeddings, for example, using some pooling layer such as a global average pooling layer. The disclosed systems can then apply the attribute localization feature embeddings to corresponding attribute group classifiers to generate tags based on predicted attributes. In particular, attribute group classifiers can predict attributes as associated with a query image (e.g., based on a scoring comparison with other potential attributes of an attribute group).

Abstract: Systems, methods, and non-transitory computer-readable media are disclosed for generating combined feature embeddings for minority class upsampling in training machine learning models with imbalanced training samples. For example, the disclosed systems can select training sample values from a set of training samples and a combination ratio value from a continuous probability distribution. Additionally, the disclosed systems can generate a combined synthetic training sample value by modifying the selected training sample values using the combination ratio value and combining the modified training sample values. Moreover, the disclosed systems can generate a combined synthetic ground truth label based on the combination ratio value. In addition, the disclosed systems can utilize the combined synthetic training sample value and the combined synthetic ground truth label to generate a combined synthetic training sample and utilize the combined synthetic training sample to train a machine learning model.

Abstract: Methods and systems are provided for facilitating the creation and utilization of a transformation function system capable of providing network agnostic performance improvement. The transformation function system receives a representation from a task neural network. The representation can be input into a composite function neural network of the transformation function system. A learned composite function can be generated using the composite function neural network. The composite function can be specifically constructed for the task neural network based on the input representation. The learned composite function can be applied to a feature embedding of the task neural network to transform the feature embedding. Transforming the feature embedding can optimize the output of the task neural network.

Abstract: Caption association techniques as part of digital content creation by a computing device are described. The computing device is configured to extract text features and bounding boxes from an input document. These text features and bounding boxes are processed to reduce a number of possible search spaces. The processing may involve generating and utilizing a language model that captures the semantic meaning of the text features to identify and filter static text, and may involve identifying and filtering inline captions. A number of bounding boxes are identified for a potential caption. The potential caption and corresponding identified bounding boxes are concatenated into a vector. The concatenated vector is used to identify relationships among the bounding boxes to determine a single bounding box associated with the caption. The determined association is utilized to generate an output digital document that includes a structured association between the caption and a data entry field.

Abstract: A standardized data model (“SDM”) includes standardized data types that indicate classifications of data elements. In a data service platform, such as a marketing data platform, a data standardization module classifies received data elements. One or more components included in the data standardization module are trained using supervised or unsupervised learning techniques to classify received data elements into a standardized data type included in the SDM. In some cases, an output of an unsupervised learning phase is provided as an input to a supervised learning phase. In some cases, a classified data element is modified by the data standardization module to indicate the standardized data type into which the data element is classified.

Abstract: A search system provides search results with images of products based on associations of primary products and secondary products from product image sets. The search system analyzes a product image set containing multiple images to determine a primary product and secondary products. Information associating the primary and secondary products are stored in a search index. When the search system receives a query image containing a search product, the search index is queried using the search product to identify search result images based on associations of products in the search index, and the result images are provided as a response to the query image.

Abstract: Techniques are disclosed for providing an interactive search session. The interactive search session is implemented using an artificial intelligence model. For example, when the artificial intelligence model receives a search query from a user, the model selects an action from a plurality of actions based on the search query. The selected action queries the user for more contextual cues about the search query (e.g., may enquire about use of the search results, may request to refine the search query, or otherwise engage the user in conversation to better understand the intent of the search). The interactive search session may be in the form, for example, of a chat session between the user and the system, and the chat session may be displayed along with the search results (e.g., in a separate section of display). The interactive search session may enable the system to better understand the user's search needs, and accordingly may help provide more focused search results.

Abstract: Embodiments of the present invention provide systems and methods for automatically generating a shoppable video. A video is parsed into one or more scenes. Products and their corresponding product information are automatically associated with the one or more scenes. The shoppable video is then generated using the associated products and corresponding product information such that the products are visible in the shoppable video based on a scene in which the products are found.

Abstract: Systems and methods are described for serving personalized content using content tagging and transfer learning. The method may include identifying content elements in an experience pool, where each of the content element is associated with one or more attribute tags, identifying a user profile comprising characteristics of a user, generating a set of user-tag affinity vectors based on the user profile and the corresponding attribute tags using a content personalization engine, generating a user-content affinity score based on the set of user-tag affinity vectors, selecting a content element from the plurality of content elements based on the corresponding user-content affinity score, and delivering the selected content element to the user.

Abstract: This disclosure relates to methods, non-transitory computer readable media, and systems that generate and render a varied-scale-topological construct for a multidimensional dataset to visually represent portions of the multidimensional dataset at different topological scales. In certain implementations, for example, the disclosed systems generate and combine (i) an initial topological construct for a multidimensional dataset at one scale and (ii) a local topological construct for a subset of the multidimensional dataset at another scale to form a varied-scale-topological construct. To identify a region from an initial topological construct to vary in scale, the disclosed systems can determine the relative densities of subsets of multidimensional data corresponding to regions of the initial topological construct and select one or more such regions to change in scale.

Abstract: Classifying structural features of a digital document by feature type using machine learning is leveraged in a digital medium environment. A document analysis system is leveraged to extract structural features from digital documents, and to classifying the structural features by respective feature types. To do this, the document analysis system employs a character analysis model and a classification model. The character analysis model takes text content from a digital document and generates text vectors that represent the text content. A vector sequence is generated based on the text vectors and position information for structural features of the digital document, and the classification model processes the vector sequence to classify the structural features into different feature types. The document analysis system can generate a modifiable version of the digital document that enables its structural features to be modified based on their respective feature types.

Abstract: An introspection network is a machine-learned neural network that accelerates training of other neural networks. The introspection network receives a weight history for each of a plurality of weights from a current training step for a target neural network. A weight history includes at least four values for the weight that are obtained during training of the target neural network up to the current step. The introspection network then provides, for each of the plurality of weights, a respective predicted value, based on the weight history. The predicted value for a weight represents a value for the weight in a future training step for the target neural network. Thus, the predicted value represents a jump in the training steps of the target neural network, which reduces the training time of the target neural network. The introspection network then sets each of the plurality of weights to its respective predicted value.

Abstract: Disclosed systems and methods generate user-session representation vectors from data generated by user interactions with online services. A transformation application executing on a computing device receives interaction data, which is generated by user devices interacting with an online service. The transformation application separates the interaction data into session datasets. The transformation involves normalizing the session datasets by modifying the rows within each session dataset by removing event identifiers and time stamps. The application transforms each normalized session dataset into a respective user-session representation vector. The application outputs the user-session representation vectors.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating trained neural network with increased robustness against adversarial attacks by utilizing a dynamic dropout routine and/or a cyclic learning rate routine. For example, the disclosed systems can determine a dynamic dropout probability distribution associated with neurons of a neural network. The disclosed systems can further drop neurons from a neural network based on the dynamic dropout probability distribution to help neurons learn distinguishable features. In addition, the disclosed systems can utilize a cyclic learning rate routine to force copy weights of a copy neural network away from weights of an original neural network without decreasing prediction accuracy to ensure that the decision boundaries learned are different.

Abstract: A conversational agent facilitates conversational searches for users. The conversational agent is a reinforcement learning (RL) agent trained using a user model generated from existing session logs from a search engine. The user model is generated from the session logs by mapping entries from the session logs to user actions understandable by the RL agent and computing conditional probabilities of user actions occurring given previous user actions in the session logs. The RL agent is trained by conducting conversations with the user model in which the RL agent selects agent actions in response to user actions sampled using the conditional probabilities from the user model.

Abstract: A digital medium environment is described to facilitate recommendations based on vectors generated using feature word embeddings. A recommendation system receives data that describes at least one attribute for a user profile, at least one item, and an interaction between the user profile and the at least one item. The recommendation system associates each user profile attribute, each item, and each interaction between a user profile and an item as a word, using natural language processing, and combines the words into sentences. The sentences are input to a word embedding model to determine feature vector representations describing relationships between the user profile attributes, items, and explicit and implicit interactions. From the feature vector representations, the recommendation system ascertains a similarity between different features.

Abstract: In some embodiments, a configuration management application accesses configuration data for a multi-target website. The configuration management application provides the user interface including a timeline area and a page display area. The timeline area is configured to display timeline entries corresponding to configurations of the multi-target website. Based on a selection of a timeline entry, the page display area is configured to display a webpage configuration corresponding to the selected timeline entry. In addition, the page display area is configured to display graphical annotations indicating interaction metrics for the configured page regions. In some cases, the timeline entries, configurations, and interaction metrics are determined based on a selection of a target segment for the multi-target website.

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media directed to navigation of a digital video. In one embodiment, a method may begin by partitioning a digital video into a number of sub-stories based at least in part on transition points identified within the digital video. The plurality of sub-stories can then be grouped into video segments based on the content of each sub-story. These video segments can then be packaged into a navigation panel in accordance with a selected template that defines a layout for the navigation panel. Such a navigation panel can present the video segments to a viewer in an interactive graphical manner that enables the viewer to navigate the one or more video segments. Other embodiments may be described and/or claimed.

Abstract: Techniques are described for machine learning-based generation of target segments is leveraged in a digital medium environment. A segment targeting system generates training data to train a machine learning model to predict strength of correlation between a set of users and a defined demographic. Further, a machine learning model is trained with visit statistics for the users to predict the likelihood that the users will visit a particular digital content platform. Those users with the highest predicted correlation with the defined demographic and the highest likelihood to visit the digital content platform can be selected and placed within a target segment, and digital content targeted to the defined demographic can be delivered to users in the target segment.

Abstract: Machine-learning based multi-step engagement strategy generation and visualization is described. Rather than rely heavily on human involvement to create delivery strategies, the described learning-based engagement system generates multi-step engagement strategies by leveraging machine-learning models trained using data describing historical user interactions with content delivered in connection with historical campaigns. Initially, the learning-based engagement system obtains data describing an entry condition and an exit condition for a campaign. Based on the entry and exit condition, the learning-based engagement system utilizes the machine-learning models to generate a multi-step engagement strategy, which describes a sequence of content deliveries that are to be served to a particular client device user (or segment of client device users).