An official website of the United States government
Fairness is becoming a rising concern in machine learning. Recent research has discovered that state-of-the-art models are amplifying social bias by making biased prediction towards some population groups (characterized by sensitive features like race or gender). Such unfair prediction among groups renders trust issues and ethical concerns in machine learning, especially for sensitive fields such as employment, criminal justice, and trust score assessment. In this paper, we introduce a new framework to improve machine learning fairness. The goal of our model is to minimize the influence of sensitive feature from the perspectives of both data input and predictive model. To achieve this goal, we reformulate the data input by eliminating the sensitive information and strengthen model fairness by minimizing the marginal contribution of the sensitive feature. We propose to learn the sensitive-irrelevant input via sampling among features and design an adversarial network to minimize the dependence between the reformulated input and the sensitive information. Empirical results validate that our model achieves comparable or better results than related state-of-the-art methods w.r.t. both fairness metrics and prediction performance.
more »
« less
My Model is Unfair, Do People Even Care? Visual Design Affects Trust and Perceived Bias in Machine Learning
Machine learning technology has become ubiquitous, but, unfortunately, often exhibits bias. As a consequence, disparate stakeholders need to interact with and make informed decisions about using machine learning models in everyday systems. Visualization technology can support stakeholders in understanding and evaluating trade-offs between, for example, accuracy and fairness of models. This paper aims to empirically answer “Can visualization design choices affect a stakeholder's perception of model bias, trust in a model, and willingness to adopt a model?” Through a series of controlled, crowd-sourced experiments with more than 1,500 participants, we identify a set of strategies people follow in deciding which models to trust. Our results show that men and women prioritize fairness and performance differently and that visual design choices significantly affect that prioritization. For example, women trust fairer models more often than men do, participants value fairness more when it is explained using text than as a bar chart, and being explicitly told a model is biased has a bigger impact than showing past biased performance. We test the generalizability of our results by comparing the effect of multiple textual and visual design choices and offer potential explanations of the cognitive mechanisms behind the difference in fairness perception and trust. Our research guides design considerations to support future work developing visualization systems for machine learning.
more »
« less
- PAR ID:
- 10515212
- Publisher / Repository:
- IEEE
- Date Published:
- Journal Name:
- IEEE transactions on visualization and computer graphics
- Volume:
- 30
- ISSN:
- 1077-2626
- Page Range / eLocation ID:
- 327 - 337
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
A gender stereotype that has some basis in research is that men are more reluctant to ask for directions than women. We wanted to investigate whether this stereotype applies to technology-related contexts, affecting older adults’ abilities to learn new technologies. To explore how help seeking and gender might relate for older adults, we conducted a controlled experiment with 36 individuals (18 men and 18 women), and observed how often they asked for help when learning new applications. We also conducted post- experiment interviews with participants. We found that although most participants stereotyped older men as being reluctant to ask for help in the interview, the difference between men and women was minimal in the experiment. Instead, other factors had a greater effect: older participants took longer to complete tasks and participants with lower technology self-efficacy asked significantly more questions.more » « less
-
With the rise of AI, algorithms have become better at learning underlying patterns from the training data including ingrained social biases based on gender, race, etc. Deployment of such algorithms to domains such as hiring, healthcare, law enforcement, etc. has raised serious concerns about fairness, accountability, trust and interpretability in machine learning algorithms. To alleviate this problem, we propose D-BIAS, a visual interactive tool that embodies human-in-the-loop AI approach for auditing and mitigating social biases from tabular datasets. It uses a graphical causal model to represent causal relationships among different features in the dataset and as a medium to inject domain knowledge. A user can detect the presence of bias against a group, say females, or a subgroup, say black females, by identifying unfair causal relationships in the causal network and using an array of fairness metrics. Thereafter, the user can mitigate bias by refining the causal model and acting on the unfair causal edges. For each interaction, say weakening/deleting a biased causal edge, the system uses a novel method to simulate a new (debiased) dataset based on the current causal model while ensuring a minimal change from the original dataset. Users can visually assess the impact of their interactions on different fairness metrics, utility metrics, data distortion, and the underlying data distribution. Once satisfied, they can download the debiased dataset and use it for any downstream application for fairer predictions. We evaluate D-BIAS by conducting experiments on 3 datasets and also a formal user study. We found that D-BIAS helps reduce bias significantly compared to the baseline debiasing approach across different fairness metrics while incurring little data distortion and a small loss in utility. Moreover, our human-in-the-loop based approach significantly outperforms an automated approach on trust, interpretability and accountability.more » « less
-
AI systems have been known to amplify biases in real-world data. Explanations may help human-AI teams address these biases for fairer decision-making. Typically, explanations focus on salient input features. If a model is biased against some protected group, explanations may include features that demonstrate this bias, but when biases are realized through proxy features, the relationship between this proxy feature and the protected one may be less clear to a human. In this work, we study the effect of the presence of protected and proxy features on participants’ perception of model fairness and their ability to improve demographic parity over an AI alone. Further, we examine how different treatments—explanations, model bias disclosure and proxy correlation disclosure—affect fairness perception and parity. We find that explanations help people detect direct but not indirect biases. Additionally, regardless of bias type, explanations tend to increase agreement with model biases. Disclosures can help mitigate this effect for indirect biases, improving both unfairness recognition and decision-making fairness. We hope that our findings can help guide further research into advancing explanations in support of fair human-AI decision-making.more » « less
-
Urban population growth has significantly complicated the management of mobility systems, demanding innovative tools for planning. Generative Crowd-Flow (GCF) models, which leverage machine learning to simulate urban movement patterns, offer a promising solution but lack sufficient evaluation of their fairness–a critical factor for equitable urban planning. We present an approach to measure and benchmark the fairness of GCF models by developing a first-of-its-kind set of fairness metrics specifically tailored for this purpose. Using observed flow data, we employ a stochastic biased sampling approach to generate multiple permutations of Origin-Destination datasets, each demonstrating intentional bias. Our proposed framework allows for the comparison of multiple GCF models to evaluate how models introduce bias in outputs. Preliminary results indicate a tradeoff between model accuracy and fairness, underscoring the need for careful consideration in the deployment of these technologies. To this end, this study bridges the gap between human mobility literature and fairness in machine learning, with potential to help urban planners and policymakers leverage GCF models for more equitable urban infrastructure development.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/TVCG.2023.3327192
