Recidivism Model

Recidivism prediction models aim to forecast the likelihood of criminal re-offense, assisting in parole decisions and resource allocation. Current research heavily emphasizes mitigating bias and promoting fairness within these models, exploring techniques like data preprocessing to remove the influence of sensitive variables and incorporating fairness-enhancing regularization into model training. This work highlights the inherent trade-offs between accuracy, fairness, and interpretability, demanding careful consideration of model selection and evaluation metrics to ensure responsible and equitable application. The ultimate goal is to develop models that are both accurate and just, minimizing discriminatory outcomes while providing valuable insights for criminal justice reform.

Papers

October 20, 2024

How Aligned are Generative Models to Humans in High-Stakes Decision-Making?
Sarah Tan, Keri Mallari, Julius Adebayo, Albert Gordo, Martin T. Wells, Kori Inkpen
Generative Model Real Human LLM Alignment Large Generative Model Risk Score Recidivism Model

October 11, 2024

JAILJUDGE: A Comprehensive Jailbreak Judge Benchmark with Multi-Agent Enhanced Explanation Evaluation Framework
Fan Liu, Yue Feng, Zhao Xu, Lixin Su, Xinyu Ma, Dawei Yin, Hao Liu
Jailbreak Attack Legal Document Multi Agent Framework Explanation Evaluation Recidivism Model Jailbreak Evaluation

March 26, 2024

Counterfactual Fairness through Transforming Data Orthogonal to Bias
Shuyi Chen, Shixiang Zhu
Absolute Stance Bias Counterfactual Fairness Biased Decision Orthogonal Transforms Recidivism Model

December 22, 2023

SoK: Taming the Triangle -- On the Interplays between Fairness, Interpretability and Privacy in Machine Learning
Julien Ferry, Ulrich Aïvodji, Sébastien Gambs, Marie-José Huguet, Mohamed Siala
Machine Learning Inherent Interpretability Procedural Fairness Decision Making Privacy Policy Responsible Machine Learning Recidivism Model

November 11, 2023

Is Machine Learning Unsafe and Irresponsible in Social Sciences? Paradoxes and Reconsidering from Recidivism Prediction Tasks
Jianhong Liu, Dianshi Li
Machine Learning Computational Approach Social Science Fermi Paradox Recidivism Model Expert Prediction

July 21, 2023

Framework for developing quantitative agent based models based on qualitative expert knowledge: an organised crime use-case
Frederike Oetker, Vittorio Nespeca, Thijs Vis, Paul Duijn, Peter Sloot, Rick Quax
Full Model New Framework Agent Based Model Crime Pattern Data Analysis Agent Recidivism Model Criminal Network

February 17, 2023

On (assessing) the fairness of risk score models
Eike Petersen, Melanie Ganz, Sune Hannibal Holm, Aasa Feragen
Procedural Fairness Algorithmic Fairness Fair Credit Credit Scoring Model Recidivism Model

January 15, 2023

Doubly Robust Counterfactual Classification
Kwangho Kim, Edward H. Kennedy, José R. Zubizarreta
High Quality Counterfactuals Doubly Robust Recidivism Model Counterfactual Classification

September 14, 2022

Accuracy, Fairness, and Interpretability of Machine Learning Criminal Recidivism Models
Eric Ingram, Furkan Gursoy, Ioannis A. Kakadiaris
Inherent Interpretability Procedural Fairness Recidivism Model

November 5, 2021

Increasing Fairness in Predictions Using Bias Parity Score Based Loss Function Regularization
Bhanu Jain, Manfred Huber, Ramez Elmasri
Loss Function Human Prediction Procedural Fairness Bias Score Fairness Loss Recidivism Model

Recidivism Model

Papers

How Aligned are Generative Models to Humans in High-Stakes Decision-Making?

JAILJUDGE: A Comprehensive Jailbreak Judge Benchmark with Multi-Agent Enhanced Explanation Evaluation Framework

Counterfactual Fairness through Transforming Data Orthogonal to Bias

SoK: Taming the Triangle -- On the Interplays between Fairness, Interpretability and Privacy in Machine Learning

Is Machine Learning Unsafe and Irresponsible in Social Sciences? Paradoxes and Reconsidering from Recidivism Prediction Tasks

Framework for developing quantitative agent based models based on qualitative expert knowledge: an organised crime use-case

On (assessing) the fairness of risk score models

Doubly Robust Counterfactual Classification

Accuracy, Fairness, and Interpretability of Machine Learning Criminal Recidivism Models

Increasing Fairness in Predictions Using Bias Parity Score Based Loss Function Regularization