Survey of technologies for trustworthy machine learning systems.
problem Building machine learning systems that are fair, explainable, auditable, and secure.
method Survey of technologies across data and model stages of machine learning.
result Four categories of system properties (fairness, explainability, auditability, safety & security) are essential for trustworthy systems.
Neural approach enhances AI trustworthiness, generalization, and robustness.
problem Challenges in explaining, generalizing, and adapting AI models to uncertain environments.
method Customized trustworthy networks, flexible learning regularizers, open-world recognition losses.
result Significant performance improvements across various open-world multimedia recognition scenarios.
Proposes a game-theoretic framework for ML trust regulation.
problem Lack of coordination between ML model builders and regulators.
method Formulates trustworthy ML as a multi-objective multi-agent optimization problem and introduces regulation games and ParetoPlay.
result Enables efficient enforcement of ML model specifications without discouraging participation.
New research finds tree-based GAMs are most trustworthy and fair.
problem Variability in GAM algorithms leads to inconsistent models.
method Quantitative and qualitative analysis of various GAM algorithms.
result Tree-based GAMs are the most trustworthy and fair.
New framework makes ML methods compliant with regulations.
problem Ensuring ML methods meet regulatory standards.
method InfoGram and Admissible Machine Learning framework.
result Redesigns ML methods for regulatory compliance.
SMILE improves explainability of machine learning models.
problem Difficulty in understanding and trusting the conclusions of black-box machine learning models.
method Statistical Model-agnostic Interpretability with Local Explanations (SMILE).
result SMILE makes machine learning models more interpretable.
In many contexts, it can be useful for domain experts to understand to what extent predictions made by a machine learning model can be trusted. In particular, estimates of trustworthiness can be useful for fraud analysts who process machine learning-generated alerts of fraudulent transactions. In this work, we present …
Research tackles distribution shift issues in ML to improve AI reliability.
problem Distribution shift limits ML reliability and trustworthiness.
method Study three distribution shifts (perturbation, domain, modality) and investigate robustness, explainability, adaptability.
result Proposes effective solutions and fundamental insights for enhancing ML robustness, adaptability, and safety.
Clarifies challenges in machine learning uncertainty quantification.
problem Inconsistent terminology and diverse technical requirements for trustworthy uncertainties.
method Examines estimation targets, uncertainty constructs, and problematic mappings.
result Advocates for alignment between intent and implementation in UQ.
Conformal prediction improves signal detection accuracy in railway images.
problem Improving the reliability of machine learning models for railway signal detection.
method Applying conformal prediction to a novel dataset of train operator perspective images.
result The approach enhances the reliability of machine learning models for detecting railway signals.
Bayesian autoencoders quantify anomaly uncertainty for safer machine learning.
problem Lack of uncertainty quantification in autoencoders for anomaly detection.
method Formulated Bayesian autoencoders to quantify epistemic and aleatoric anomalies.
result Demonstrated effectiveness of BAEs on benchmark and real datasets.
Paper explores differential privacy in high-dimensional federated learning, tackling server trustworthiness and estimation.
problem Maintaining privacy in distributed environments with high-dimensional data.
method Investigates scenarios with untrusted and trusted central servers, introduces novel federated estimation algorithms for linear regression models.
result Tight minimax rates depend on high-dimensionality even with sparsity assumptions, and novel algorithms handle slight variations among distributed models.
Study identifies negative data externalities affecting model performance on specific groups.
problem Negative data externalities on group performance in machine learning models.
method Characterized and detected data-model inefficiencies, focusing on specific types of externalities.
result Negative data externalities can lower model performance on specific sub-groups, even with larger datasets.
Bayesian meta learning improves uncertainty quantification in regression.
problem Trusting uncertainty quantification in Bayesian regression.
method Trust-Bayes framework for Bayesian meta learning, optimizing for trustworthy uncertainty quantification.
result Lower bounds and sample complexity for trustworthy uncertainty quantification are characterized.
The Intensive Care Unit (ICU) is a hospital department where machine learning has the potential to provide valuable assistance in clinical decision making. Classical machine learning models usually only provide point-estimates and no uncertainty of predictions. In practice, uncertain predictions should be presented to …
This work improves neural network trustworthiness through uncertainty estimation.
problem Overconfident neural networks lead to poor performance under distribution shifts.
method Develops a general uncertainty framework for neural networks, including classification with rejection.
result Improves model trustworthiness and robustness in decision-making tasks.
Paper tackles selective regression using uncertainty estimation.
problem Selective regression for machine learning models to avoid predictions when uncertain.
method Model-agnostic non-parametric uncertainty estimation.
result Superior performance compared to state-of-the-art selective regressors.
Enhancement attacks can falsely improve machine learning model performance in biomedical research.
problem The trustworthiness of machine learning in biomedical research is threatened by enhancement attacks.
method Developed two techniques to enhance prediction performance with minimal changes to features.
result Falsely improved classifiers' accuracy from 50% to almost 100% while maintaining high feature similarities.
While social networks can provide an ideal platform for up-to-date information from individuals across the world, it has also proved to be a place where rumours fester and accidental or deliberate misinformation often emerges. In this article, we aim to support the task of making sense from social media data, and speci…
ICA model detects misinformation during high impact events like COVID-19.
problem Detecting misinformation during high impact events like the COVID-19 pandemic.
method Data-driven solution based on ICA model.
result ICA model achieves knowledge discovery and detection of misinformation jointly.
Proposes a new trust framework for AI models to maximize utility.
problem Concerns over bias and discrimination in predictive models.
method Introduces a novel trust framework inspired by philosophy, focusing on maximizing Bayes utility.
result Properly-ranked models are inherently U-trustworthy. AI needs causal inference to avoid being just a correlation machine.
problem AI's inability to distinguish correlation from causation.
method Develops a unified framework connecting various causal statistical estimators and proves a Statistical Necessity Theorem for causal generalization.
result AI systems without causal grounding are brittle and biased, highlighting the need for causal statistics.
A new score function improves explainability and reliability of AI systems.
problem Designing AI systems that are explainable, robust, and trustworthy.
method Integrates conformal prediction with explainable machine learning using a novel score function.
result The method achieves improved performance on target classes and satisfies conformal guarantees.
Optimizes calibration error estimators for better classifier trustworthiness.
problem Lack of guidance on selecting and tuning calibration error estimators.
method Reformulates calibration estimation as a regression problem with i.i.d. input pairs.
result Demonstrates the effectiveness of optimized calibration estimators on image classification tasks.
A framework assesses the trustworthiness of probabilistic classifiers using local calibration error.
problem Assessing the trustworthiness of probabilistic classifiers beyond traditional metrics.
method I-trustworthy framework linking local calibration to trustworthiness; Kernel Local Calibration Error (KLCE) method for hypothesis testing.
result The effectiveness of the proposed test statistic demonstrated through simulated and real-world datasets.
Bayesian framework improves ML classification models' uncertainty estimates.
problem Ensuring trustworthy AI predictions with explicit uncertainty quantification.
method Proposes a Bayesian framework for generative ML classification models that accounts for input measurement uncertainty.
result The BQDA model outperforms other models in terms of interpretability, explicit uncertainty modeling, and computational efficiency.
Study shows trust and trustworthiness emerge through reinforcement learning.
problem Trust and trustworthiness are universal but not predicted by traditional economic models.
method Used Q-learning algorithm to simulate trust and trustworthiness dynamics in a trust game.
result High levels of trust and trustworthiness emerge when individuals consider both past and future experiences.
Online knowledge repositories typically rely on their users or dedicated editors to evaluate the reliability of their content. These evaluations can be viewed as noisy measurements of both information reliability and information source trustworthiness. Can we leverage these noisy evaluations, often biased, to distill a…
A framework for analyzing regularizers to ensure trustworthy theory-driven model estimation.
problem Uncertain choice of regularizers can compromise the interpretability of deep grey-box models.
method Adapting neural net architecture and training objective to analyze regularizer behavior empirically.
result Empirical analysis of regularizers helps in making a justified choice for trustworthy theory-driven model estimation.
Efficient approach improves prediction calibration for domain shifts.
problem Improving uncertainty-aware predictions for domain shifts.
method Combining entropy-encouraging and adversarial calibration losses.
result Substantially outperforms existing approaches in domain drift calibration.
New report on machine learning visualization techniques and trends.
problem Improving trust in machine learning models through visualization.
method Analysis of peer-reviewed articles on machine learning visualization techniques.
result Rapid growth in machine learning visualization techniques over the past three years.
We demonstrate how easy it is for modern machine-learned systems to violate common deontological ethical principles and social norms such as "favor the less fortunate," and "do not penalize good attributes." We propose that in some cases such ethical principles can be incorporated into a machine-learned model by adding…
Adaptive Quantum Conformal Prediction improves reliability of quantum machine learning predictions.
problem Quantum machine learning lacks robust uncertainty quantification methods.
method Adaptive Conformal Inference applied to quantum conformal prediction to maintain validity over time.
result AQCP achieves target coverage levels and is more stable than standard quantum conformal prediction.
Federated Learning solves privacy and data distribution challenges in machine learning.
problem Privacy concerns and distributed data in machine learning.
method Federated Learning framework to handle privacy and distributed data.
result Solves privacy and data distribution challenges in machine learning.
This thesis enhances ML reliability by selectively abstaining from predictions when uncertain.
problem Improving reliability in machine learning systems, especially in high-stakes domains.
method Exploiting uncertainty signals from training trajectories to develop lightweight, post-hoc abstention methods compatible with differential privacy.
result A robust trajectory-based approach to selective prediction that maintains high accuracy under privacy noise.
Diamond method controls FDR for trustworthy feature interaction discovery in ML models.
problem Limited interpretability of ML models due to black box nature.
method Diamond method integrates model-X knockoffs framework to control FDR for non-additive interactions.
result Diamond method ensures accurate discovery of feature interactions with FDR control.
Most machine learning classifiers give predictions for new examples accurately, yet without indicating how trustworthy predictions are. In the medical domain, this hampers their integration in decision support systems, which could be useful in the clinical practice. We use a supervised learning approach that combines E…
Fairness in machine learning increases privacy risks, especially for underrepresented groups.
problem Privacy risks in fair machine learning models, particularly for underrepresented groups.
method Membership inference attacks to measure information leakage and analyze fairness vs. privacy trade-offs.
result Achieving fairness in machine learning models increases privacy risks, especially for underrepresented groups.
Adversarial Robustness Toolbox (ART) is a Python library supporting developers and researchers in defending Machine Learning models (Deep Neural Networks, Gradient Boosted Decision Trees, Support Vector Machines, Random Forests, Logistic Regression, Gaussian Processes, Decision Trees, Scikit-learn Pipelines, etc.) agai…
This paper reviews counterfactual explanations for machine learning models.
problem Making machine learning models understandable to humans.
method Categorization and evaluation of counterfactual explanation algorithms.
result A rubric for evaluating counterfactual explanations.
The paper advocates for interpretable, accountable, reproducible machine learning in medicine.
problem Black box models in medicine lack transparency and regulatory approval.
method Intrinsically interpretable modeling approaches and collaborative learning paradigms.
result Interpretable machine learning models can support clinical decisions and gain regulatory approval.
ManifoldMind uses adaptive-curvature probabilistic spheres for trustworthy recommendations in semantic hierarchies.
problem Sparse and abstract recommendation domains where users explore diverse conceptual paths.
method Adaptive-curvature probabilistic spheres, soft multi-hop inference, and curvature-aware semantic kernel.
result Superior NDCG, calibration, and diversity compared to baselines on public benchmarks.
Paper explores Rashomon set models for more trustworthy medical conclusions.
problem Lack of comprehensive analysis of models in Rashomon set leads to misleading conclusions.
method Introduces Rashomon_DETECT algorithm and Profile Disparity Index (PDI).
result Combining differently behaving models in Rashomon set provides more trustworthy conclusions.
The paper connects three machine learning methods to reduce generalization errors.
problem Reducing generalization errors in machine learning models.
method Distributionally robust optimization, Bayesian methods, and regularization.
result Machine learning models can be characterized using distributional uncertainty and robustness measures.
Online health communities are a valuable source of information for patients and physicians. However, such user-generated resources are often plagued by inaccuracies and misinformation. In this work we propose a method for automatically establishing the credibility of user-generated medical statements and the trustworth…
AI systems need reliable testing to ensure safety and trustworthiness.
problem Current AI Act lacks functional trustworthiness for AI systems.
method Define technical application distribution, set risk-based performance, and conduct statistically valid testing.
result Reliable functional trustworthiness is essential for AI systems.
Fairness is becoming a rising concern w.r.t. machine learning model performance. Especially for sensitive fields such as criminal justice and loan decision, eliminating the prediction discrimination towards a certain group of population (characterized by sensitive features like race and gender) is important for enhanci…
Consensus dimension reduction combines multiple visualizations to identify shared patterns.
problem Conflicting visualizations from different dimension reduction methods.
method Multi-view learning to identify stable patterns across multiple views.
result Consensus visualization effectively identifies shared low-dimensional data structure.