Paper uses conformal prediction sets to make criminal justice risk assessments fairer.
problem Fairness issues in criminal justice risk assessment algorithms.
method Adopting conformal prediction sets to remove unfairness from algorithms and covariates.
result Constructs confusion tables and measures fairness effectively free of racial differences.
Study examines how people perceive fairness in criminal risk prediction algorithms.
problem Concerns about fairness in algorithmic decision making, especially in criminal risk prediction.
method Survey of 576 people to understand perceptions of fairness in algorithmic decision making.
result People's fairness judgments are influenced by eight latent properties of features in algorithms.
Machine learning predicts criminal networks' missing partnerships and future behavior.
problem Predicting and understanding criminal networks' properties and future behavior.
method Combining graph representation learning and machine learning methods.
result Outstanding accuracy in predicting missing criminal partnerships and future behavior.
Network analysis helps prevent money laundering by identifying risky clients and suspicious clusters.
problem Preventing money laundering using social network analysis.
method Real-world data analysis, network metrics, predictive models, visual analysis.
result Risk profiles can be predicted using social network metrics.
Objectives: Discussions of fairness in criminal justice risk assessments typically lack conceptual precision. Rhetoric too often substitutes for careful analysis. In this paper, we seek to clarify the tradeoffs between different kinds of fairness and between fairness and accuracy. Methods: We draw on the existing liter…
New method reduces unfairness in binary classification.
problem Achieving similar false positive and negative rates across two populations.
method Penalizes unfairness to achieve balanced false positive and negative rates.
result Empirically validated approach improves fairness and accuracy.
Recidivism prediction instruments provide decision makers with an assessment of the likelihood that a criminal defendant will reoffend at a future point in time. While such instruments are gaining increasing popularity across the country, their use is attracting tremendous controversy. Much of the controversy concerns …
Paper introduces xAUC metric to assess fairness of risk scores in bipartite ranking tasks.
problem Disparate impact of risk scores in non-binary, downstream uses.
method Investigates fairness in bipartite ranking tasks, introduces xAUC metric.
result xAUC metric reveals disparities not seen in binary classification performance.
Recidivism prediction instruments (RPI's) provide decision makers with an assessment of the likelihood that a criminal defendant will reoffend at a future point in time. While such instruments are gaining increasing popularity across the country, their use is attracting tremendous controversy. Much of the controversy c…
DeFi doesn't fully remove trust, showing run risk and personal character's importance.
problem The need for trust in DeFi despite its code-based approach.
method Natural experiment revealing identities of DeFi participants, including a criminal.
result DeFi doesn't fully remove trust, showing run risk and personal character's relevance.
Risk scores are simple classification models that let users make quick risk predictions by adding and subtracting a few small numbers. These models are widely used in medicine and criminal justice, but are difficult to learn from data because they need to be calibrated, sparse, use small integer coefficients, and obey …
New model detects hidden group structures in criminal networks.
problem Challenges in identifying group structures in criminal networks with noisy data.
method Developed an extended stochastic block model (ESBM) to infer group structures.
result Unveiled complex block structures in an Italian mafia network.
New fair-by-design model reduces bias in recidivism prediction.
problem Discriminatory bias in recidivism prediction models.
method Prototype-based, locally learned, data distribution extraction.
result Reduces bias and provides interpretable rules.
Interpretable ML models predict recidivism as well as non-interpretable methods and are more fair.
problem Improving recidivism prediction models for fairness and interpretability.
method Trained interpretable ML models on two recidivism datasets, compared to existing methods, and analyzed fairness.
result Interpretable ML models can predict recidivism as well as non-interpretable methods and are more fair.
New framework makes ML methods compliant with regulations.
problem Ensuring ML methods meet regulatory standards.
method InfoGram and Admissible Machine Learning framework.
result Redesigns ML methods for regulatory compliance.
Paper predicts crimes using historical data and machine learning.
problem Predicting and preventing crime activities.
method Supervised learning with decision tree, k-nearest neighbor, Random Forest, and Adaboost algorithms.
result The system predicts crimes with better accuracy using historical data.
Method interprets deep learning models using topological data analysis.
problem Lack of interpretability in deep learning models, especially in high-risk applications.
method Topological and geometric data analysis to infer features and decision-making mechanisms of DL models.
result Extracted subgraphs reveal relevant features for model decisions, demonstrating model's reliance on pertinent data.
Paper presents mdfa to identify victims of discrimination in black box classifiers.
problem Identifying victims of discrimination in black box classifiers.
method Reduces discrimination measurement to matching distributions and sensitive attribute coincidence prediction.
result Identifies African-American individuals at high risk of violent recidivism.
Bayesian model assesses criminal threat escalation.
problem Evaluate threats posed by potential violent criminals.
method Customised three-level Bayesian hierarchical model.
result Model calibrates to expert judgments and updates in real time.
New metrics improve fairness in risk assessments.
problem Risk assessments reflect historical policies, not future decisions.
method Counterfactual analogues of metrics, doubly robust estimation.
result Fairness metrics under counterfactuals can differ from standard metrics.
Predictive modeling is increasingly being employed to assist human decision-makers. One purported advantage of replacing human judgment with computer models in high stakes settings-- such as sentencing, hiring, policing, college admissions, and parole decisions-- is the perceived "neutrality" of computers. It is argued…
Study disrupts Sicilian Mafia networks using data analysis.
problem Challenges in disrupting resilient criminal networks.
method Social Network Analysis methods applied to real-world datasets.
result Neutralizing only 5% of affiliates can disrupt network connectivity by 70%.
This thesis evaluates the quality of binary classification models in crime forecasting tools.
problem Quality assessment of binary classification models in crime forecasting tools.
method Binary classifier approach to evaluate AUC and PPVk. result The PPVk deviates significantly from AUC in crime forecasting models. Neural network model predicts alternating event-free periods.
problem Dynamic prediction of alternating recurrent events with statistical nuance.
method Developed an online dynamic prediction framework using neural network theory.
result Outstanding performance in predicting alternating recurrent event-free time.
The paper reconciles two conflicting fairness criteria in algorithmic risk scores.
problem How to reconcile calibration and equal error rates in algorithmic risk scores.
method Derive necessary and sufficient conditions for existence of calibrated scores achieving equal error rates, then present an algorithm to find the most accurate score subject to both criteria.
result The method can eliminate error disparities while maintaining calibration and improve profit in credit lending.
New approach to algorithmic fairness for human-AI collaboration considers compliance with human decisions.
problem Current fairness approaches assume perfect human compliance, but real-world compliance is often poor.
method Defines compliance-robustly fair algorithms and proposes an optimization strategy to improve fairness.
result Algorithmic recommendations can improve fairness even if humans do not fully comply with fair algorithms.
New method improves fairness in biased predictions.
problem Improving fairness in biased classifier predictions.
method Individual bias detector prioritizes data samples for a bias mitigation algorithm.
result Superior performance in individual and group fairness on real-world datasets.
Avoid explaining black box models; use interpretable ones instead.
problem High-stakes decisions made by black box models cause societal harm.
method Design inherently interpretable models instead of trying to explain black box models.
result Inherently interpretable models are a better approach for high-stakes decisions.
Develops a method to quantify racial bias in law enforcement systems.
problem Quantify racial bias in law enforcement systems considering criminality and multi-stage interactions.
method Multi-stage causal framework incorporating criminality.
result Identifies three canonical scenarios of racial bias in law enforcement.
The study improves crime prediction using Foursquare and streetlight data with demographic info.
problem Improving crime prediction models using diverse data sources.
method Combining Foursquare and streetlight data with demographic info for crime prediction.
result The proposed model enhances classification performance in crime prediction.
Improves pre-trial risk assessments by making them safer without changing existing rules.
problem Improving pre-trial risk assessments while maintaining deterministic rules.
method Developed a maximin robust optimization approach to find a safer policy.
result Can safely improve certain components of the risk assessment instrument.
Inspection-L detects illicit cryptocurrency transactions using GNNs and self-supervised learning.
problem Detect illicit cryptocurrency transactions for anti-money laundering.
method Graph Neural Network (GNN) framework based on self-supervised Deep Graph Infomax (DGI) and Graph Isomorphism Network (GIN) with supervised learning algorithms.
result Inspection-L outperforms state-of-the-art methods in key classification metrics.
The paper argues that fairness in predictions should be evaluated in context and addressed through data collection.
problem Fairness in predictive models in sensitive applications like healthcare and criminal justice.
method Decompose cost-based metrics of discrimination into bias, variance, and noise; propose actions to estimate and reduce each term; perform case-studies.
result Data collection is often a means to reduce discrimination without sacrificing accuracy.
FairPOT balances fairness and AUC performance by selectively transforming risk scores.
problem Balancing fairness and AUC performance in high-stakes domains.
method FairPOT uses proportional optimal transport to selectively transform risk scores.
result FairPOT consistently improves fairness with minimal AUC degradation or even positive gains.
This monograph introduces deep learning models for predicting time-to-event outcomes.
problem Predicting critical events and their timing from time series data.
method Neural networks and deep learning models for survival analysis.
result Improved accuracy in predicting time-to-event outcomes using deep learning.
Study identifies and measures biases in legal case data.
problem Addressing representation biases and sentencing disparities in legal case data.
method Utilizes two regression models: a baseline and a fair judge model.
result Quantifies biases across demographic groups in criminal data from Cook County (Illinois).
Local discovery method uncovers direct unfairness in complex systems.
problem Identifying causal pathways of unfairness in complex domains.
method Local discovery for direct discrimination (LD3) method.
result LD3 returns a valid adjustment set (VAS) for assessing unfairness.
Introduces Rashomon Capacity to measure predictive multiplicity in probabilistic classifiers.
problem Predictive multiplicity in classification models leading to unjustified decisions.
method Introduces Rashomon Capacity, a metric for probabilistic classifiers, and provides a rigorous derivation.
result Rashomon Capacity captures nuanced score variations and provides strategies for disclosing conflicting models.
The paper introduces a model comparison framework for identifying fairness differences between machine learning models.
problem Comparing fairness in machine learning models across different subgroups.
method Develops a framework to automatically identify subgroups where models differ in fairness metrics.
result Identifies subgroups where machine learning models disagree on fairness-related quantities.
Fairness measures fail in predictive settings that intentionally shift outcomes.
problem Fairness measures fail in performative prediction settings.
method Formalized concept shift and counterfactual outcomes.
result Predictors that are fair during training become unfair during deployment.
Algorithm learns fair allocations in unknown group frequencies.
problem Fairness in resource allocation problems.
method Efficient learning algorithm in a censored feedback model.
result Algorithm converges to optimal fair allocation.
The paper studies and mitigates accuracy disparity in regression models.
problem Accuracy disparity between different demographic subgroups in high-stakes domains.
method Error decomposition theorem and distribution alignment algorithm.
result The proposed algorithm effectively mitigates accuracy disparity while maintaining predictive power.
New framework improves fairness in machine learning models.
problem Eliminating discrimination in machine learning models for sensitive fields.
method Reformulate data input by removing sensitive features and design adversarial network to minimize dependence.
result Our model achieves better fairness metrics and prediction performance than state-of-the-art methods.
Novel framework for Bayesian neural networks incorporating task-specific constraints.
problem Task-specific constraints in supervised model deployment.
method Introduces Output-Constrained BNN (OC-BNN) framework.
result OC-BNNs effectively incorporate prior expert knowledge and desiderata like safety and fairness.
New approach to fairness in machine learning through stochastic optimization.
problem Fairness issues in machine learning predictions.
method Stochastic multi-objective optimization for accuracy-fairness trade-offs.
result Well-spread and accurate Pareto fronts for handling streaming data.
This research proposes a variable importance cloud to assess variable importance across multiple good models.
problem Current variable importance measures are tied to a single model, limiting understanding of variable importance across different models.
method Introduces a variable importance cloud that maps every variable to its importance for every good predictive model.
result Shows how variable importance can vary significantly across different good models.
Gamification enhances law enforcement training on terrorism financing.
problem Complex technologies limit engagement in fighting cyber-criminal activities.
method Combining learning and training methods with gamification.
result Participants show increased expertise and positive feedback.
Discussing the need for explainable AI in various fields.
problem The lack of transparency in AI and ML methods.
method Discussion of explainable AI from a grounded perspective.
result Highlighting the importance of explainable AI in fields like health and justice.