Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: The present disclosure provides computing systems and associated methods for optimizing one or more adjustable parameters (e.g. operating parameters) of a system. In particular, the present disclosure provides a parameter optimization system that can perform one or more black-box optimization techniques to iteratively suggest new sets of parameter values for evaluation. The iterative suggestion and evaluation process can serve to optimize or otherwise improve the overall performance of the system, as evaluated by an objective function that evaluates one or more metrics. The present disclosure also provides a novel black-box optimization technique known as “Gradientless Descent” that is more clever and faster than random search yet retains most of random search's favorable qualities.

Abstract: The present disclosure provides computing systems and associated methods for optimizing one or more adjustable parameters (e.g. operating parameters) of a system. In particular, the present disclosure provides a parameter optimization system that can perform one or more black-box optimization techniques to iteratively suggest new sets of parameter values for evaluation. The iterative suggestion and evaluation process can serve to optimize or otherwise improve the overall performance of the system, as evaluated by an objective function that evaluates one or more metrics. The present disclosure also provides a novel black-box optimization technique known as “Gradientless Descent” that is more clever and faster than random search yet retains most of random search's favorable qualities.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: Methods and systems for performing black box optimization to identify an output that optimizes an objective.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: A computer-implemented method can include receiving, by one or more computing devices, one or more prior evaluations of performance of a machine learning model, the one or more prior evaluations being respectively associated with one or more prior variants of the machine-learning model, the one or more prior variants of the machine-learning model each having been configured using a different set of adjustable parameter values. The method can include utilizing, by the one or more computing devices, an optimization algorithm to generate a suggested variant of the machine-learning model based at least in part on the one or more prior evaluations of performance and the associated set of adjustable parameter values, the suggested variant of the machine-learning model being defined by a suggested set of adjustable parameter values.

Abstract: A computer-implemented method is provided for optimization of parameters of a system, product, or process. The method includes establishing an optimization procedure for a system, product, or process. The system, product, or process has an evaluable performance that is dependent on values of one or more adjustable parameters. The method includes receiving one or more prior evaluations of performance of the system, product, or process. The one or more prior evaluations are respectively associated with one or more prior variants of the system, product, or process. The one or more prior variants are each defined by a set of values for the one or more adjustable parameters. The method includes utilizing an optimization algorithm to generate a suggested variant based at least in part on the one or more prior evaluations of performance and the associated set of values.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium, including a method for providing content. A set of k requests is identified, each request being associated with a request for content. Candidate content items are identified for one or more of the requests in the set. Identified candidate content items are ranked based at least in part on their respective associated bids, expected value or long-term value. Top K candidate content items are determined based on the ranking. A price is assigned to each of the top K candidate content items based on the ranking and the associated bids. One or more of the top K candidate content items are provided responsive to the requests in the set.

Abstract: Methods, and systems, including computer programs encoded on computer-readable storage mediums, including a method for providing to a user device a rule set specifying adjustments to be made by the user device to predicted performance measures of content items presented by the user device. The method includes receiving a content item request from a user device for content items, the user device being associated with a user device identifier; determining content items responsive to the content item request, each of the content items have a predicted performance measure; identifying a rule set specifying adjustments to be performed by the user device to the predicted performance measures of the determined content items in response to particular user interactions with the determined content items when presented on the user device; and providing the determined content items and the rule set to the user device in response to the content item request.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for managing decision models. In one aspect, a method includes providing data that cause presentation of a model development user interface receiving first model rule data through the user interface. The first model rule data specify a first characteristic of a violating resource and a threshold score for the first characteristic. Additional model rule data are received through the user interface. The additional model rule data specify one or more additional model rules. Relationship data are received through the user interface for each of the additional model rules. The relationship data specify sets of the additional model rules that violating resources satisfy. Data that cause a hierarchical presentation of the first model rule and the additional model rules are provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for analyzing content item compliance with specified guidelines. In one aspect, a method includes receiving training data that specify manual classifications of content items and feature values for each of the content items, where each manual classification specifies whether the content item is a violating content item. Using the training data, a compliance model is trained to classify an unclassified content item as a violating content item based on the feature values of the unclassified content item. A determination is made that the compliance model has an accuracy measure that meets a threshold accuracy measure. In response to determining that the accuracy measure for the compliance model meets the accuracy threshold, unclassified content items are classified using the feature values for the unclassified content items, and data specifying the classifications are provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for outcome estimation and content item selection. In one aspect, a method includes training an outcome estimation model using both regression prediction quality measures and ranking prediction quality measures. In response to receiving a request for one or more content items; a predicted outcome is computed for each content item in a set of content items, each respective outcome being computed using the outcome estimation model and feature values associated with each respective the content item. Selection scores are computed for the content items using the predicted outcome. In turn, one or more content items are selected for presentation based on the selection scores; and data are provided that cause presentation of the one or more content items at a client device.