Patents Assigned to Fairness-as-a-Service, Inc.
  • Systems and methods for intelligent generation and assessment of candidate less discriminatory alternative machine learning models
    Patent number: 12248858
    Abstract: A system and method includes obtaining an incumbent model and a candidate model, generating a plurality of synthetic model input datasets, computing, for each synthetic model input dataset, a model performance efficacy metric and a model fairness efficacy metric for the incumbent model based on assessing model output data of the incumbent model that corresponds to each respective synthetic model input dataset of the plurality of synthetic model input datasets, computing, for each synthetic model input dataset, a model performance efficacy metric and a model fairness efficacy metric for the candidate model based on assessing model output data of the candidate model that corresponds to each respective synthetic model input dataset of the plurality of synthetic model input datasets, computing, for the candidate model, a disparity-mitigating model viability score, and displaying, via a graphical user interface, a representation of the candidate model in association with the disparity-mitigating model viability sc
    Type: Grant
    Filed: June 12, 2024
    Date of Patent: March 11, 2025
    Assignee: Fairness-as-a-Service, Inc.
    Inventors: John Wickens-Lamb Merrill, Kareem Saleh, Mark Eberstein
  • SYSTEMS AND METHODS FOR INTELLIGENT GENERATION AND ASSESSMENT OF CANDIDATE LESS DISCRIMINATORY ALTERNATIVE MACHINE LEARNING MODELS
    Publication number: 20240428133
    Abstract: A system and method includes obtaining an incumbent model and a candidate model, generating a plurality of synthetic model input datasets, computing, for each synthetic model input dataset, a model performance efficacy metric and a model fairness efficacy metric for the incumbent model based on assessing model output data of the incumbent model that corresponds to each respective synthetic model input dataset of the plurality of synthetic model input datasets, computing, for each synthetic model input dataset, a model performance efficacy metric and a model fairness efficacy metric for the candidate model based on assessing model output data of the candidate model that corresponds to each respective synthetic model input dataset of the plurality of synthetic model input datasets, computing, for the candidate model, a disparity-mitigating model viability score, and displaying, via a graphical user interface, a representation of the candidate model in association with the disparity-mitigating model viability sc
    Type: Application
    Filed: June 12, 2024
    Publication date: December 26, 2024
    Applicant: Fairness-as-a-Service, Inc.
    Inventors: John Wickens-Lamb Merrill, Kareem Saleh, Mark Eberstein
  • Disparity mitigation in machine learning-based predictions for distinct classes of data using derived indiscernibility constraints during neural network training
    Patent number: 12039457
    Abstract: A system and method includes generating approximate distributions for distinct classes of data samples; computing a first partial Jensen-Shannon (JS) divergence and a second partial JS divergence based on the approximate distribution of the disparity affected class of data samples with reference to the approximate distribution of the control class of data samples; computing a disparity divergence based on the first partial JS divergence and the second partial JS divergence; generating a distribution-matching term based on the disparity divergence, wherein the distribution-matching term mitigates an inferential disparity between the control class of data samples and the disparity affected class of data samples during a training of an unconstrained artificial neural network; constructing a disparity-constrained loss function based on augmenting a target loss function with the distribution-matching term; and transforming the unconstrained ANN to a disparity-constrained ANN based on a training of the unconstraine
    Type: Grant
    Filed: December 27, 2023
    Date of Patent: July 16, 2024
    Assignee: Fairness-as-a-Service, Inc.
    Inventors: John Wickens-Lamb Merrill, Kareem Saleh, Mark Eberstein
  • DISPARITY MITIGATION IN MACHINE LEARNING-BASED PREDICTIONS FOR DISTINCT CLASSES OF DATA USING DERIVED INDISCERNIBILITY CONSTRAINTS DURING NEURAL NETWORK TRAINING
    Publication number: 20240135186
    Abstract: A system and method includes generating approximate distributions for distinct classes of data samples; computing a first partial Jensen-Shannon (JS) divergence and a second partial JS divergence based on the approximate distribution of the disparity affected class of data samples with reference to the approximate distribution of the control class of data samples; computing a disparity divergence based on the first partial JS divergence and the second partial JS divergence; generating a distribution-matching term based on the disparity divergence, wherein the distribution-matching term mitigates an inferential disparity between the control class of data samples and the disparity affected class of data samples during a training of an unconstrained artificial neural network; constructing a disparity-constrained loss function based on augmenting a target loss function with the distribution-matching term; and transforming the unconstrained ANN to a disparity-constrained ANN based on a training of the unconstraine
    Type: Application
    Filed: December 27, 2023
    Publication date: April 25, 2024
    Applicant: Fairness-as-a-Service, Inc.
    Inventors: John Wickens-Lamb Merrill, Kareem Saleh, Mark Eberstein
  • DISPARITY MITIGATION IN MACHINE LEARNING-BASED PREDICTIONS FOR DISTINCT CLASSES OF DATA USING DERIVED INDISCERNIBILITY CONSTRAINTS DURING NEURAL NETWORK TRAINING
    Publication number: 20230267334
    Abstract: A system and method includes generating approximate distributions for distinct classes of data samples; computing a first partial Jensen-Shannon (JS) divergence and a second partial JS divergence based on the approximate distribution of the disparity affected class of data samples with reference to the approximate distribution of the control class of data samples; computing a disparity divergence based on the first partial JS divergence and the second partial JS divergence; generating a distribution-matching term based on the disparity divergence, wherein the distribution-matching term mitigates an inferential disparity between the control class of data samples and the disparity affected class of data samples during a training of an unconstrained artificial neural network; constructing a disparity-constrained loss function based on augmenting a target loss function with the distribution-matching term; and transforming the unconstrained ANN to a disparity-constrained ANN based on a training of the unconstraine
    Type: Application
    Filed: May 1, 2023
    Publication date: August 24, 2023
    Applicant: Fairness-as-a-Service, Inc.
    Inventors: John Wickens-Lamb Merrill, Kareem Saleh, Mark Eberstein
  • Disparity mitigation in machine learning-based predictions for distinct classes of data using derived indiscernibility constraints during neural network training
    Patent number: 11676037
    Abstract: A system and method includes generating approximate distributions for distinct classes of data samples; computing a first partial Jensen-Shannon (JS) divergence and a second partial JS divergence based on the approximate distribution of the disparity affected class of data samples with reference to the approximate distribution of the control class of data samples; computing a disparity divergence based on the first partial JS divergence and the second partial JS divergence; generating a distribution-matching term based on the disparity divergence, wherein the distribution-matching term mitigates an inferential disparity between the control class of data samples and the disparity affected class of data samples during a training of an unconstrained artificial neural network; constructing a disparity-constrained loss function based on augmenting a target loss function with the distribution-matching term; and transforming the unconstrained ANN to a disparity-constrained ANN based on a training of the unconstraine
    Type: Grant
    Filed: December 5, 2022
    Date of Patent: June 13, 2023
    Assignee: Fairness-as-a-Service, Inc.
    Inventors: John Wickens-Lamb Merrill, Kareem Saleh, Mark Eberstein