"Why Should I Trust You?"

Sameer Singh; Carlos Guestrin; Marco Túlio Ribeiro

doi:10.1145/2939672.2939778

"Why Should I Trust You?"

Почему я должен вам доверять?

Sameer Singh, Carlos Guestrin, Marco Túlio Ribeiro

2016-08-08

SCID: 54.1/m5w6g3zc Discuss with AI

black-box modelsmodel interpretabilitymodel transparencyprediction explanationstrustworthy machine learning

Abstract (AI)

Despite widespread adoption, machine learning models remain mostly black boxes. Understanding the reasons behind predictions is, however, quite important in assessing trust, which is fundamental if one plans to take action based on a prediction, or when choosing whether to deploy a new model. Such understanding also provides insights into the model, which can be used to transform an untrustworthy model or prediction into a trustworthy one.

Key Findings

Explanations of predictions can provide insights into model behavior and help identify ways to transform untrustworthy models or predictions into trustworthy ones.

The abstract highlights the opacity of widely adopted machine-learning models as a central challenge to trust and deployment decisions.

Understanding the reasons behind machine-learning predictions is important for assessing trust before acting on predictions or deploying models.

Research Object

Machine learning models and their predictions

Research Subject

Interpretability of model predictions as a basis for assessing and improving trustworthiness and deployment decisions

Publication Details

DOI

10.1145/2939672.2939778

Publication Date

2016-08-08

Journal

Publisher

ISSN

Access Type

Author Information

Authors

Sameer Singh

Carlos Guestrin

Marco Túlio Ribeiro

Download PDF

Make a presentation

References available in scid.ai2

Scikit-learn: Machine Learning in Python2012 Least angle regression2004

Cited by20

Review of deep learning: concepts, CNN architectures, challenges, applications, future directions2021 Interpreting Black-Box Models: A Review on Explainable Artificial Intelligence2023 Explainable Artificial Intelligence (XAI): What we know and what is left to attain Trustworthy Artificial Intelligence2023 A Survey on the Explainability of Supervised Machine Learning2021 A Unifying Review of Deep and Shallow Anomaly Detection2021 FinBERT : A Large Language Model for Extracting Information from Financial Text*2022 Connecting the dots in trustworthy Artificial Intelligence: From AI principles, ethics, and key requirements to responsible AI systems and regulation2023 Evaluating the Quality of Machine Learning Explanations: A Survey on Methods and Metrics2021 Financial Fraud: A Review of Anomaly Detection Techniques and Recent Advances2021 From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI2023 Explainability in Graph Neural Networks: A Taxonomic Survey2022 Explainable Artificial Intelligence (XAI) techniques for energy and power systems: Review, challenges and opportunities2022 Explainable Artificial Intelligence Applications in Cyber Security: State-of-the-Art in Research2022 Artificial Intelligence for Predictive Maintenance Applications: Key Components, Trustworthiness, and Future Trends2024 Interpreting artificial intelligence models: a systematic review on the application of LIME and SHAP in Alzheimer’s disease detection2024 Recognition of Sunflower Diseases Using Hybrid Deep Learning and Its Explainability with AI2023 Machine learning methods in macroeconomic forecasting: Preliminary results2025 Machine learning evaluation of structural descriptors for supercooled water2026 A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI2020 Legal and human rights issues of AI: Gaps, challenges and vulnerabilities2020