Efficient DNN-Powered Software with Fair Sparse Models

2024-07-03 · Xuanqi Gao, Weipeng Jiang, Juan Zhai, Shiqing Ma, XiaoYu Zhang, Chao Shen

With the emergence of the Software 3.0 era, there is a growing trend of compressing and integrating large models into software systems, with significant societal implications. Regrettably, in numerous instances, model compression techniques impact the fairness performance of these models and thus the ethical behavior of DNN-powered software. One of the most notable example is the Lottery Ticket Hypothesis (LTH), a prevailing model pruning approach. This paper demonstrates that fairness issue of LTHbased pruning arises from both its subnetwork selection and training procedures, highlighting the inadequacy of existing remedies. To address this, we propose a novel pruning framework, Ballot, which employs a novel conflict-detection-based subnetwork selection to find accurate and fair subnetworks, coupled with a refined training process to attain a high-performance model, thereby improving the fairness of DNN-powered software. By means of this procedure, Ballot improves the fairness of pruning by 38.00%, 33.91%, 17.96%, and 35.82% compared to state-of-the-art baselines, namely Magnitude Pruning, Standard LTH, SafeCompress, and FairScratch respectively, based on our evaluation of five popular datasets and three widely used models. Our code is available at https://anonymous.4open.science/r/Ballot-506E.

📄 PDF Abstract BibTeX arXiv:2407.02805

Code (0)

등록된 구현이 없습니다.

Tasks

Fairness Model Compression

Methods 이 논문이 사용한 방법론

Pruning 설명 없음

Data Collection and Quality Challenges in Deep Learning: A Data-Centric AI Perspective

2021-12-13 · Steven Euijong Whang, Yuji Roh, Hwanjun Song, Jae-Gil Lee

Data-centric AI is at the center of a fundamental shift in software engineering where machine learning becomes the new software, powered by big data and computing infrastructure. Here software engineering needs to be re-…

BIG-bench Machine Learning Fairness Feature Engineering Management

Astraea: Grammar-based Fairness Testing

2020-10-06 · Ezekiel Soremekun, Sakshi Udeshi, Sudipta Chattopadhyay

Software often produces biased outputs. In particular, machine learning (ML) based software are known to produce erroneous predictions when processing discriminatory inputs. Such unfair program behavior can be caused by …

Fairness Fault Diagnosis

Software Engineering for Fairness: A Case Study with Hyperparameter Optimization

2019-05-14 · Joymallya Chakraborty, Tianpei Xia, Fahmid M. Fahid, Tim Menzies

We assert that it is the ethical duty of software engineers to strive to reduce software discrimination. This paper discusses how that might be done. This is an important topic since machine learning software is increasi…

BIG-bench Machine Learning Fairness Hyperparameter Optimization

Designing Adaptive Digital Nudging Systems with LLM-Driven Reasoning

2026-04-13 · Tiziano Santilli, Mina Alipour, Mahyar Tourchi Moghaddam arxiv

Digital nudging systems lack architectural guidance for translating behavioral science into software design. While research identifies nudge strategies and quality attributes, existing architectures fail to integrate mul…

Towards Fair Machine Learning Software: Understanding and Addressing Model Bias Through Counterfactual Thinking

2023-02-16 · Zichong Wang, Yang Zhou, Israat Haque, David Lo 외

The increasing use of Machine Learning (ML) software can lead to unfair and unethical decisions, thus fairness bugs in software are becoming a growing concern. Addressing these fairness bugs often involves sacrificing ML…

Benchmarking counterfactual Fairness