FAIR ISAAC CORPORATION

In transactional systems where past transactions can have impact on the current score of a machine learning based decision model, the transactions that are most responsible for the score and the associated reasons are determined by the transactional system. A system and method identifies such past transactions that maximally impact the current score and allow for a more effective understanding of the scores generated by a model in a transactional system and explanation of specific transactions for automated decisioning, to explain the scores in terms of past transactions. Further an existing instance-based explanation system is used to identify the reasons for the score, and how the identified transactions influence these reasons. A combination of impact on score and impact on reasons determines the most impactful past transaction with respect to the most recent score being explained.

A method includes determining, by a trained machine learning model, a score based at least on one or more latent features. The method also includes monitoring the determining of the score by the trained machine learning model. The monitoring includes determining one or more production statistics associated with the one or more latent features, derived variables and input data elements, and accessing one or more reference assets persisted on a model governance blockchain. The one or more reference assets includes one or more reference statistics and a threshold indicating a deviation between the one or more production statistics and the one or more reference statistics. The method also includes generating an alert based on the one or more production statistics associated with the one or more latent features meeting the threshold. Related methods and articles of manufacture are also disclosed.

A method for dynamic data analysis and segmentation for performance monitoring within a population, the method comprising: maintaining, by a processor, a dataset comprising score distributions and odds-to-score relationships for a population over multiple time periods; identifying, by the processor, a plurality of segments within the population based on predefined characteristics stored in a database, wherein the segments are updated in real-time based on predefined rules; executing, by the processor, a machine learning (ML) model trained to detect shifts associated with each segment by analyzing changes in the score distribution or odds-to-score relationships over the multiple time periods; and generating, by the processor, segment profiles that include statistical measures for each identified segment based on the detected shifts, wherein the statistical measures include at least a mean score, a median score, and changes in the odds-to-score relationship.

In one aspect there is provided a method. The method may include collecting one or more functions that implement the decision logic of a solution. A snapshot of the one or more functions can be generated. The snapshot can executable code associated with the one or more functions. The solution can be deployed by at least storing the snapshot of the one or more functions to a repository Systems and articles of manufacture, including computer program products, are also provided.

Explanatory dropout systems and methods for improving a computer implemented machine learning model are provided using on-manifold/on-distribution evaluation of dropout of key features to explain model outputs. The machine learning model is trained using a plurality of input examples, including input records with explicit dropout operators applied effectuating the removal of influence of features associated with an explanation reason class. One or more dropout operators may be stochastically applied to one or more input examples. The procedure includes on-manifold/on-distribution evaluation of the machine learning model under conditions of absence or presence of the one or more dropout operators for reliable calculation of numerical statistics associated with reason classes to yield model explanations. The training and evaluation procedures present advantages over traditional off-manifold or off-distribution perturbative explanation procedures.

A system and method for analyzing coverage, bias and model explanations in large dimensional modeling data includes discretizing three or more variables of a dataset to generate a discretized phase space represented as a grid of a plurality of cells, the dataset comprising a plurality of records, each record of the plurality of records having a value and a unique identifier (ID). A grid transformation is applied to each record in the dataset to assign each record to a cell of the plurality of cells of the grid according to the grid transformation. A grid index is generated to reference each cell using a discretized feature vector. A grid storage for storing the records assigned to each cell of the grid is then created. The grid storage using the ID of each record as a reference to each record and the discretized feature vector as a key to each cell.

A method for generating a classifier, comprising: initializing a neural network with a fully connected architecture; applying a regularization constraint to the set of weights between neurons in the input layer and the plurality of latent features in the hidden layer; iteratively reducing a number of incoming connections to each latent feature in the hidden layer to a predetermined number based on the regularized first set of weights; weighting loss function's value based on categories of activation tuples, wherein contributions of data entries with activation tuples of size greater than a predetermined threshold to the loss function's value are limited and wherein contributions of data entries with activation tuples of size “0” to the loss function's value are minimized, to a predefined percentage; and selectively updating sets of weights associated with top-ranked latent features as evaluated by the magnitude of their contributions at the output layer in a training process.

A method for generating a classifier, wherein the method comprises generating a set of candidate latent features from a training dataset, evaluating, by the at least one processor, each of the candidate latent features based on a coverage efficiency metric, wherein the coverage efficiency metric balances detection of minority class instances against a minimization of false positives among majority class instances; selecting, by the at least one processor, a first latent feature from the set of candidate latent features based on a ranking of the coverage efficiency metric associated with each of the candidate latent features; partitioning the training dataset by marking the training records identified by the first latent feature as covered records and removing the covered records from the training dataset to form an uncovered dataset; identifying slack features from remaining candidate latent features, training a neural network model using the selected latent features and the slack features.

Computer-implemented machines, systems and methods for generating a confidence interval, in response to determining an uncertainty value associated with a first machine learning model output; switching from the first machine learning model to a second machine learning model, in response to determining the uncertainty value meets a threshold, wherein the second machine learning model generates a second machine learning model output; and providing to a user interface, the first machine learning output, the uncertainty value, the confidence interval, and the second machine learning output. The confidence interval may be represented as [max({tilde over (x)}−f(c)s, 0 ), min({tilde over (x)}+f(c)s, 1 )], where c is first a desired confidence level, {tilde over (x)} represents sample scores sample mean, s represents the sample standard deviation, and f(c) represents an appropriate parametric multiplier.

Systems and methods for training a machine learning model implemented over a network configured to represent the machine learning model are provided. At least one or more directed edges connect the one or more nodes with an edge representing a connection between a first node and a second node, the second node computing an activation depending on the values of activations on first nodes and values associated with the connections, the connection being either conforming or non-conforming. The machine learning model may be trained by iteratively adjusting parameters w and b, respectively associated with weights and biases associated with edges connecting computational nodes. Connections between nodes may be sparsified by adjusting the parameter w to a first value for non-conforming connections during the training phase to reduce complexity of the connections among the plurality of nodes, or to ensure the input-output function of the network adheres to additional constraints.