Table of Links Abstract and Introduction


Related work


Methods
3.1 Positive-sum fairness
3.2 Application


Experiments
4.1 Initial results
4.2 Positive-sum fairness


Conclusion and References 4.1 Initial results According to traditional group fairness, in assessing the results of the four models shown in figure 3a one could conclude that: M2 improves the overall performance Our results show that M2 outperforms M1 in terms of AUROC. This is in line with our expectation as we are providing an additional relevant medical feature for the model to learn from. This better performance comes with a larger gap in AUROC between the most advantaged and most discriminated races, in other words less fairness from a traditional point of view. But this larger disparity is not necessarily harmful according to the positive-sum fairness as we will discuss it in the next section. M4 improves the fairness M4 improves fairness for lung lesions and consolidations, while performing similar for pneumonia and pleural effusion. The improved fairness is likely due to the gradient reversal layer, which removes race information from the image and prevents the model from exploiting any demographic shortcut. No clear pattern for M3 The results for M3 are less consistent. Its performance is lower than the baseline except for pneumonia and its fairness measurement is sometimes higher and other times lower than the baseline’s. If the baseline model exploited demographic encodings present in the images to generate shortcuts, training M3 to maximize the race prediction might have intensified the impact of these shortcuts. Authors:
(1) Samia Belhadj∗, Lunit Inc., Seoul, Republic of Korea (samia.belhadj@lunit.io);
(2) Sanguk Park [0009 −0005 −0538 −5522]*, Lunit Inc., Seoul, Republic of Korea (tony.superb@lunit.io);
(3) Ambika Seth, Lunit Inc., Seoul, Republic of Korea (ambika.seth@lunit.io);
(4) Hesham Dar [0009 −0003 −6458 −2097], Lunit Inc., Seoul, Republic of Korea (heshamdar@lunit.io);
(5) Thijs Kooi [0009 −0003 −6458 −2097], Kooi, Lunit Inc., Seoul, Republic of Korea (tkooi@lunit.io). This paper is available on arxiv under CC BY-NC-SA 4.0 license. Table of Links Abstract and Introduction Related work Methods
3.1 Positive-sum fairness
3.2 Application Experiments
4.1 Initial results
4.2 Positive-sum fairness Conclusion and References Abstract and Introduction Abstract and Introduction Abstract and Introduction Related work Related work Related work Methods 3.1 Positive-sum fairness 3.2 Application Methods 3.1 Positive-sum fairness 3.1 Positive-sum fairness 3.2 Application 3.2 Application Experiments 4.1 Initial results 4.2 Positive-sum fairness Experiments Experiments 4.1 Initial results 4.1 Initial results 4.2 Positive-sum fairness 4.2 Positive-sum fairness Conclusion and References Conclusion and References Conclusion and References 4.1 Initial results According to traditional group fairness, in assessing the results of the four models shown in figure 3a one could conclude that: M2 improves the overall performance Our results show that M2 outperforms M1 in terms of AUROC. This is in line with our expectation as we are providing an additional relevant medical feature for the model to learn from. This better performance comes with a larger gap in AUROC between the most advantaged and most discriminated races, in other words less fairness from a traditional point of view. But this larger disparity is not necessarily harmful according to the positive-sum fairness as we will discuss it in the next section. M2 improves the overall performance M4 improves the fairness M4 improves fairness for lung lesions and consolidations, while performing similar for pneumonia and pleural effusion. The improved fairness is likely due to the gradient reversal layer, which removes race information from the image and prevents the model from exploiting any demographic shortcut. M4 improves the fairness No clear pattern for M3 The results for M3 are less consistent. Its performance is lower than the baseline except for pneumonia and its fairness measurement is sometimes higher and other times lower than the baseline’s. If the baseline model exploited demographic encodings present in the images to generate shortcuts, training M3 to maximize the race prediction might have intensified the impact of these shortcuts. No clear pattern for M3 Authors: (1) Samia Belhadj∗, Lunit Inc., Seoul, Republic of Korea (samia.belhadj@lunit.io); (2) Sanguk Park [0009 −0005 −0538 −5522]*, Lunit Inc., Seoul, Republic of Korea (tony.superb@lunit.io); (3) Ambika Seth, Lunit Inc., Seoul, Republic of Korea (ambika.seth@lunit.io); (4) Hesham Dar [0009 −0003 −6458 −2097], Lunit Inc., Seoul, Republic of Korea (heshamdar@lunit.io); (5) Thijs Kooi [0009 −0003 −6458 −2097], Kooi, Lunit Inc., Seoul, Republic of Korea (tkooi@lunit.io). Authors: Authors: (1) Samia Belhadj∗, Lunit Inc., Seoul, Republic of Korea (samia.belhadj@lunit.io); (2) Sanguk Park [0009 −0005 −0538 −5522]*, Lunit Inc., Seoul, Republic of Korea (tony.superb@lunit.io); (3) Ambika Seth, Lunit Inc., Seoul, Republic of Korea (ambika.seth@lunit.io); (4) Hesham Dar [0009 −0003 −6458 −2097], Lunit Inc., Seoul, Republic of Korea (heshamdar@lunit.io); (5) Thijs Kooi [0009 −0003 −6458 −2097], Kooi, Lunit Inc., Seoul, Republic of Korea (tkooi@lunit.io). This paper is available on arxiv under CC BY-NC-SA 4.0 license. This paper is available on arxiv under CC BY-NC-SA 4.0 license. available on arxiv

Part of HackerNoon's growing list of open-source research papers, promoting free access to academic material.

How AI Models Could Detect Lung Conditions Fairly

New Study Shows How Positive-Sum Fairness Impacts Medical AI Models in Chest Radiography

Read My Stories

Too Long; Didn't Read

Get Free Unlimited IP Geo Data With IPinfo Lite

New Findings Show How Positive-Sum Fairness Changes the Performance of Medical AI Models

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

Untitled Story

Achieving Fair AI Without Sacrificing Accuracy

Pay attention to that man behind the curtain

AI Is Inherently Neutral - It Is Human Beings Who Are Biased, and the Machines Merely Replicate Them

AI Is Not the Concern - It’s AI Developers You Should Be Worried About

Are Smart Cities a Threat to Data Privacy?

Achieving Fair AI Without Sacrificing Accuracy

Pay attention to that man behind the curtain

AI Is Inherently Neutral - It Is Human Beings Who Are Biased, and the Machines Merely Replicate Them

AI Is Not the Concern - It’s AI Developers You Should Be Worried About

Are Smart Cities a Threat to Data Privacy?

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps