Abstract: A method for performing a multi-horizon forecast is provided, and may include: first feature scoring first information including past targets, past covariates, and static information, the first featuring scoring providing a relevance score for at least some features considered by the first feature scoring; past learning at least some results of the first feature scoring; time projecting at least some results of the first feature scoring; second feature scoring second information including future covariates, the second feature scoring providing a relevance score for at least some features considered by the second feature scoring; future learning a combination of at least some results of the time projecting and the second feature scoring; merging at least some results of the future learning and the past learning; outputting a forecast based on a least some results of the merging. The relevance scores provide interpretability for the forecast.

Abstract: A system is presented that implements an electronic network reconfiguration tool that improves electronic network resource usage efficiency within an electronic network. The system may be configured to: utilize perturbation theory, to generate a reconfiguration guide from a reconfiguration dataset of the electronic network, by perturbing a plurality of network configurations and a plurality of usage factors of electronic network resource usage attributes; determine, by evaluating a first set of electronic network resource usage changes against the reconfiguration guide, a set of network configuration changes that improves the electronic network resource usage efficiency; and improve the electronic network resource usage efficiency by implementing the set of network configuration changes within the electronic network.

Abstract: A method and a system for assessing trustworthiness of an artificial intelligence (AI) model are provided. The method includes: accessing an explainable artificial intelligence (XAI) asset that relates to the AI model; accessing a document that relates to the AI model; separating the XAI asset into XAI asset chunks; separating the document into document chunks; generating, via a pre-trained model, a plurality of chunk embeddings that includes respective embeddings for the XAI asset chunks and the document chunks; accessing a query that relates to the trustworthiness of the AI model; generating, via the pre-trained model, a query embedding for the query; calculating an embedding similarity score for each chunk embedding; selecting a chunk embedding having the highest embedding similarity score; generating a context embedding by aggregating the query embedding with the selected chunk embedding; and generating a response to the query based on the context embedding.

Abstract: A method and a system for assessing trustworthiness of an artificial intelligence (AI) model are provided. The method includes: accessing an explainable artificial intelligence (XAI) asset that relates to the AI model; accessing a document that relates to the AI model; separating the XAI asset into XAI asset chunks; separating the document into document chunks; generating, via a pre-trained model, a plurality of chunk embeddings that includes respective embeddings for the XAI asset chunks and the document chunks; accessing a query that relates to the trustworthiness of the AI model; generating, via the pre-trained model, a query embedding for the query; calculating an embedding similarity score for each chunk embedding; selecting a chunk embedding having the highest embedding similarity score; generating a context embedding by aggregating the query embedding with the selected chunk embedding; and generating a response to the query based on the context embedding.

Abstract: A method and system for generating cluster level explanations for an input data having limited or special values are disclosed. The method includes obtaining an input data set and a clustering stopping criteria, the input data set including multiple features, and each of the features having multiple feature values; performing features decomposition on the input data; performing correlation analysis based on the features decomposition and a correlation threshold value; grouping the features having different feature values into multiple clusters; training a model based on the obtained input data; obtaining a target input data to be tested and a number of cores; grouping the number of cores into multiple clusters; and computing explanations at a cluster level.

Abstract: Aspects of the subject disclosure may include, for example, systems and methods for generating structured datasets for predicting bond price. The systems and methods include constructing a price function of each bond contained in a plurality of bond clusters including a target cluster, training a machine learning model to determine a cause for an erroneous price prediction result, and generating structured datasets based on a feedback from the machine learning model. Other embodiments are disclosed.

Abstract: Various methods and processes, apparatuses/systems, and media for generating recourse data for a denied applicant are disclosed.

Abstract: A method and system for retrieving targeted information from a document by a large language model (LLM). The method includes receiving a document and a query for targeted information; tagging the document with sentence tags; splitting the tagged document into first and second segments; implementing a LLM; and assigning the first segment and the second segment to the LLM in a chronological order. The method also includes instructing the LLM to identify and select a first set of relevant tokens within the first segment and identify and select a second set of relevant tokens within the second segment, performing a prompt-based approach or an attention-based approach that highlights relevant tokens by the LLM from the first set of relevant tokens and from the second set of relevant tokens, and providing an output of the targeted information from the LLM.

Abstract: Methods and systems for obtaining contextual information about a machine learning model are provided. The method includes: receiving raw data that is usable for training a model; training the by using the raw data; computing a set of common background data based on the raw data; computing a first explanation based on an output of the model and the set of common background data; computing, based on an output of the model, an agnostic model representation of the model; computing, based on the first explanation and the agnostic model representation, a deep, compact, and dense explanation-driven representation of the model; and determining, based on the explanation-driven representation, contextual information that relates to the model.

Abstract: Various methods and processes, apparatuses or systems, and media for performing fairness aware optimization are disclosed. The present disclosure provides acquiring and quantizing a plurality of input features for an application, and binning of each of the plurality of input features quantized so that each of the plurality of input features is assigned to a bin among a plurality of bins to provide a matrix of bin membership for each of the plurality of applications. A vector is then generated based on the matrix of bin membership for evaluating the generated vector against a target value and presence of disparity in outcome for a protected input feature. Once a coefficient vector that optimizes an output while negating any disparity in outcome for the protected input feature is identified, an optimization model is updated with such coefficient vector for subsequent processing.

Abstract: Various methods and processes, apparatuses/systems, and media for generating recourse data with data-driven actionability constraints for a negatively classified individual are disclosed. A processor trains a machine learning model by using a first set of training data and a second set of training data which outputs risk classification data associated with a negative decision; identifies, based on the risk classification data, a negatively classified individual who received the negative decision; applies a feature attribution algorithm to the trained model; ranks, in response to applying the feature attribution algorithm, a list of features that explain a negative classification for the negatively classified individual; filters the list of features that explain the negative classification for each negatively classified individual by utilizing computed actionability labels (i.e.

Abstract: A method and system for mitigating predictive multiplicity in a gradient boosting model (GBM). The method includes generating an empirical parameter set based on an approximation search resulting in a subset that includes candidates of at least one weak learner model (WLM) from a predetermined set of WLMs and training iteratively the empirical parameter set to derive a group filtered from the subset based on at least one from among a model selection (MS) technique and an intermediate ensembles (IE) technique. The method also includes selecting sequentially the at least one WLM from the derived group based on the at least one from among the MS technique and the IE technique; and generating the GBM based on a compilation of the sequentially selected at least one WLM, wherein the generated GBM operates below a minimum predefined disagreement threshold related to assessing predictive multiplicity, thereby mitigating the predictive multiplicity.

Abstract: A method and a system for determining a recourse path with respect to a decision that is associated with a positive outcome and a negative outcome are provided. The method includes: receiving a dataset including a data point representing an entity; determining, via a trained model, which data points from the dataset reach a positive outcome and which data points reach a negative outcome based on a distance threshold; determining transition labels from historical data; calculating an optimal distance function and an optimal threshold value for the dataset based on the transition labels; generating an augmentation algorithm based on the optimal distance function and the optimal threshold value; and generating a first recourse path for the first entity to reach the positive outcome by applying the augmentation algorithm to insert a second data point into the at least one dataset.

Abstract: A method and system for detecting harmful shift in a machine learning (ML) model associated with unlabeled data utilized by the ML model. The method includes implementing an error estimator model with regressor algorithm and training the error estimator model with a first portion of a labeled calibration dataset. The method further includes computing, by the trained error estimator model, an error estimation threshold based on a second portion of the labeled calibration dataset; predicting a performance of the ML model by detecting the harmful shift via the trained error estimator model analyzing the unlabeled data over a predetermined time period and determining a proportion of estimated errors associated with the unlabeled data over the predetermined time period that exceeds the error estimation threshold; and generate an alert when the proportion of estimated errors exceeds the error estimation threshold.

Abstract: Various methods and processes, apparatuses or systems, and media for mitigating disparities between outputs of different AI/ML models that are subject to usage constraints are disclosed. The method includes: receiving uncertainty values and model quality-related parameter values that are associated with at least two models that are configured to generate a loan price for a loan applicant; receiving feature weight functions that relate to weights of target metrics; calculating model weights for each model; selecting a customized model based on the model weights; receiving a tabular set of personal data that includes individualized financial information and individualized demographic information associated with loan applicants; training the customized model by using the tabular set of personal data, the target metrics, a set of sensitive labels, and historical information that relates to outputs generated by the models; and using the trained model to generate a customized loan price for a loan applicant.

Abstract: Various methods, apparatuses/systems, and media for computing change-agnostic data points are disclosed. A processor trains a machine learning model by using the at least the first set of raw data; computes a set of explanations for all combinations based on output data of the trained machine learning model, the first set of raw data, and sampled raw data computed by applying a sampling algorithm on the raw data; computes a compact representation of the set of explanations corresponding to a pre-configured dimension based on compression quality and generating a set of compressed explanations; computes a unique representation of model explanation with respect to the pre-configured dimension; determines whether the model explanation is robust to changes in data through data perturbation; and computes change-agnostic data points based on determining that the model explanation is robust to changes in data through data perturbation.

Abstract: Various methods, apparatuses/systems, and media for computing strategies for model inferences are disclosed. A processor generates background data from raw data and data sampling strategies associated with a particular security instrument; computes a model for each pair of raw data and machine learning algorithm; computes model explanation for each pair of the background data and the model; normalizes the computed model explanation by utilizing a predefined algorithm; computes a deep dense representation of each explanation based on the normalized explanation of the computed model; clusters the deep dense representation of each explanation; computes a deep dense representation of explanation of each predicted target data version associated with features recovery; compares the deep dense representation of each explanation with the deep dense representation of explanation of each predicted target data version; and computes a strategy of model selection for each target data version as output.

Abstract: Various methods and processes, apparatuses or systems, and media for computing a fair market value of a bond are disclosed. A processor generates a table where all weight vector associated with a pricing prediction value of the bond at a given time received from a plurality of data sources are included therein; receives weight vector as input corresponding to the bond from the table; and computes a loss function for each of the plurality of data sources individually, wherein each loss function includes a first part and a second part, the first term indicates a distance very closer to a real value of the price at which the trade was executed compared to the second part which is a term that penalizes predictions for being further away from the real value; and computes a fair market value of the bond based on the loss function.

Abstract: 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.

Abstract: A method for computing regional counterfactual rules to summarize recourse options is disclosed. The method includes receiving model data via an input, the model data including information that relates to a target model and a corresponding data set; training a surrogate model for the target model based on the model data; identifying, by using the surrogate model, rules for an output of the target model based on a predetermined threshold, the rules including a counterfactual rule; enumerating, by using the surrogate model, boundaries for the identified rules to partition an input space into a grid with cells, each of the cells including a hyperrectangular cell; labeling, by using the surrogate model, each of the cells with one of the identified rules based on predetermined optimality criteria; and merging, by using the surrogate model, each of the labeled cells based on a matching of the rules to generate regions of optimality.