A Conceptual Framework for Ethical Evaluation of Machine Learning Systems
August 05, 2024 Β· Declared Dead Β· π AAAI/ACM Conference on AI, Ethics, and Society
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Neha R. Gupta, Jessica Hullman, Hari Subramonyam
arXiv ID
2408.10239
Category
cs.CY: Computers & Society
Cross-listed
cs.AI,
cs.LG,
cs.SE
Citations
5
Venue
AAAI/ACM Conference on AI, Ethics, and Society
Last Checked
5 months ago
Abstract
Research in Responsible AI has developed a range of principles and practices to ensure that machine learning systems are used in a manner that is ethical and aligned with human values. However, a critical yet often neglected aspect of ethical ML is the ethical implications that appear when designing evaluations of ML systems. For instance, teams may have to balance a trade-off between highly informative tests to ensure downstream product safety, with potential fairness harms inherent to the implemented testing procedures. We conceptualize ethics-related concerns in standard ML evaluation techniques. Specifically, we present a utility framework, characterizing the key trade-off in ethical evaluation as balancing information gain against potential ethical harms. The framework is then a tool for characterizing challenges teams face, and systematically disentangling competing considerations that teams seek to balance. Differentiating between different types of issues encountered in evaluation allows us to highlight best practices from analogous domains, such as clinical trials and automotive crash testing, which navigate these issues in ways that can offer inspiration to improve evaluation processes in ML. Our analysis underscores the critical need for development teams to deliberately assess and manage ethical complexities that arise during the evaluation of ML systems, and for the industry to move towards designing institutional policies to support ethical evaluations.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
π Similar Papers
In the same crypt β Computers & Society
π
π
The Cartographer
R.I.P.
π»
Ghosted
Artificial Intelligence: the global landscape of ethics guidelines
R.I.P.
π»
Ghosted
The role of artificial intelligence in achieving the Sustainable Development Goals
R.I.P.
π»
Ghosted
Green AI
R.I.P.
π»
Ghosted
Principles alone cannot guarantee ethical AI
R.I.P.
π»
Ghosted
Tackling Climate Change with Machine Learning
Died the same way β π» Ghosted
R.I.P.
π»
Ghosted
Federated Learning: Strategies for Improving Communication Efficiency
R.I.P.
π»
Ghosted
In-Datacenter Performance Analysis of a Tensor Processing Unit
R.I.P.
π»
Ghosted
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning
R.I.P.
π»
Ghosted