Jpmorgan Chase Bank, N.a.

Daniele Magazzeni

Daniel Borrajo

Freddy Lecue

Alberto Pozanco

Parisa Zehtabi

Michael Cashmore

Sameena Shah

Manuela Veloso

Maria Manuela Veloso

Salwa Husam Alamir

Sanjay Kariyappa

Jie Deng

Yufei Wu

Allan Anzagira

Saumitra Mishra

Danial Dervovic

Rui Manuel Ramos Teixeira Da Silva

Leonidas Tsepenekas

Methods and systems for using a non-linear machine learning model to generate explanations that relate to decisions made by the model and for repairing such explanations in order to improve quality and accuracy of model outputs are provided. The method includes: receiving a data set that corresponds to attributes that pertain to a decision to be made; inputting the data set to a machine learning model; generating a baseline decision that corresponds to an output of the model with respect to data set; computing, based on the baseline decision, an explanation that relates to at least one of the attributes; estimating one or more errors associated with the explanation; and computing, based on the estimated error(s), at least one repair that corresponds to a modification of the explanation, and a cost for repairing the explanation.

Method and system for repairing explanations for non-linear machine learning models

A method for evaluating classification models that are trained by using incomplete data sets for which data is known to be missing and the missing data is known to be non-random is provided. The method includes: receiving first information that relates to data to be used for training and evaluating a performance of a classification model that is designed to make a determination with respect to a particular query; analyzing the first information to determine second information that relates to a known portion of the first information and third information that relates to missing data; estimating, based on the third information, an uncertainty that corresponds to the missing data; and calculating, based on the second information and the estimated uncertainty, a first Gaussian approximation to a performance metric that relates to the first classification model.

Method and system for evaluation of classification models

Methods and systems for detecting attempted manipulation of a machine learning model via explanation poisoning are provided. The method includes: computing explanations based on outputs of the model that include information that relates to features that affect the output of the model with respect to the first data point; assigning labels to the explanations based on the features; generating an explanation ensemble that resides in an N-dimensional space, N being equal to a number of assigned labels plus one; determining a region within the N-dimensional space for which a subsequent introduction of data causes a subsequent explanation that does not relate to the features; and when the additional data is introduced to the determined region, generating an alert message for notifying a user that a likelihood of adverse manipulation of the model is high based on the additional data.

Method and system for detecting model manipulation through explanation poisoning

Methods and systems for performing a combinatorial optimization task are provided. The method includes: receiving a first set of data items and a second set of data items; dividing the second set of data items into a first subset and a second subset; generating, based on the first subset, a first set of subset sums, and storing the first set of subset sums as a list; generating, based on the second subset, a second set of subset sums, and storing the second set of subset sums as a hash-map in conjunction with a corresponding hash function; and detecting, based on the first set of data items, the list, and the hash-map, at least one match between a subset sum that corresponds to the second set of data items and a subset sum that corresponds to the first set of data items.

Method and system for solving subset sum matching problem using search approach

A method for optimizing personnel utilization is provided. The method includes: accessing first task management planning information that relates to a first project that has not been completed; using the accessed first task management planning information to identify a plurality of tasks to be performed in connection with the first project and to identify a plurality of persons to be assigned to respective tasks; retrieving historical task management information that relates to at least one project that has been completed; and adjusting at least a first portion of the first task management planning information based on the retrieved historical task management information.

Method and system for optimization of task management issue planning

Growing community of inventors

London, United Kingdom

Daniele Magazzeni