Abstract: A computer system may include a processor and a memory coupled thereto. The memory may include a database. The processor may be configured to randomly split the database into sub-databases and applying a database query to the sub-databases. The processor may also be configured to generate respective estimated query response values for each sub-database based upon applying the database query, calculate a median of the estimated query response values, and generate a probability distribution based upon the estimated query response values and the calculated median. The processor may further be configured to select a final estimated query response value based upon the probability distribution.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for training a machine learning model. The method includes obtaining a training data set comprising a plurality of training examples; determining i) a stochastic gradient descent step size schedule, ii) a stochastic gradient descent noise schedule, and iii) a stochastic gradient descent batch size schedule, wherein the stochastic gradient descent batch size schedule comprises a sequence of varying batch sizes; and training a machine learning model on the training data set, comprising performing stochastic gradient descent according to the i) stochastic gradient descent step size schedule, ii) stochastic gradient descent noise schedule, and iii) stochastic gradient descent batch size schedule to adjust a machine learning model loss function.

Abstract: A computer system may include a processor and a memory coupled thereto. The memory may include a database. The processor may be configured to randomly split the database into sub-databases and applying a database query to the sub-databases. The processor may also be configured to generate respective estimated query response values for each sub-database based upon applying the database query, calculate a median of the estimated query response values, and generate a probability distribution based upon the estimated query response values and the calculated median. The processor may further be configured to select a final estimated query response value based upon the probability distribution.

Abstract: A computing device determines one or more first degree correlations based on valuation information. The computing device determines one or more important variables based on at least a comparison between the absolute value of the one or more first degree correlations to a first threshold value. The computing device determines a valuation function based on at least one or more of the determined one or more important variables.

Abstract: A computing device determines one or more first degree correlations based on valuation information. The computing device determines one or more important variables based on at least a comparison between the absolute value of the one or more first degree correlations to a first threshold value. The computing device determines a valuation function based on at least one or more of the determined one or more important variables.

Abstract: A method for refining a dictionary for information extraction, the operations including: inputting a set of extracted results from execution of an extractor comprising the dictionary on a collection of text, wherein the extracted results are labeled as correct results or incorrect results; processing the extracted results using an algorithm configured to set a score of the extractor above a score threshold, wherein the score threshold balances a precision and a recall of the extractor; and outputting a set of candidate dictionary entries corresponding to a full set of dictionary entries, wherein the candidate dictionary entries are candidates to be removed from the dictionary based on the extracted results.

Abstract: A method for refining a dictionary for information extraction, the operations including: inputting a set of extracted results from execution of an extractor comprising the dictionary on a collection of text, wherein the extracted results are labeled as correct results or incorrect results; processing the extracted results using an algorithm configured to set a score of the extractor above a score threshold, wherein the score threshold balances a precision and a recall of the extractor; and outputting a set of candidate dictionary entries corresponding to a full set of dictionary entries, wherein the candidate dictionary entries are candidates to be removed from the dictionary based on the extracted results.

Abstract: A method for refining a dictionary for information extraction, the operations including: inputting a set of extracted results from execution of an extractor comprising the dictionary on a collection of text, wherein the extracted results are labeled as correct results or incorrect results; processing the extracted results using an algorithm configured to set a score of the extractor above a score threshold, wherein the score threshold balances a precision and a recall of the extractor; and outputting a set of candidate dictionary entries corresponding to a full set of dictionary entries, wherein the candidate dictionary entries are candidates to be removed from the dictionary based on the extracted results.

Abstract: A method and system for identifying calls in a Java package whose targets are guaranteed to belong to the package. According to the method an inheritance graph and access permissions of respective components in the package are determined, both of which are used in combination with the knowledge that the package is seared and signed to determine whether all the targets of a call are guaranteed to belong to the package. The identification of calls according to the invention can be performed at the time the package is sealed and signed or as a separate phase thereafter and allows for better compiler optimization.