Bayesian optimization outperformed random search in machine learning hyperparameter tuning challenge.
problem Optimizing hyperparameters of machine learning models using derivative-free methods.
method Bayesian optimization vs. random search on real datasets.
result Bayesian optimization significantly outperformed random search in held-out objective functions.
New method defends machine learning models from black-box attacks.
problem Protecting machine learning models from black-box transfer attacks.
method Randomized ensemble technique with provable security guarantee.
result Empirical evidence of security for random binary classifiers.
Avoid explaining black box models; use interpretable ones instead.
problem High-stakes decisions made by black box models cause societal harm.
method Design inherently interpretable models instead of trying to explain black box models.
result Inherently interpretable models are a better approach for high-stakes decisions.
Matched Machine Learning combines machine learning and matching for causal inference.
problem Non-interpretable methods for causal inference.
method Combines machine learning and matching for interpretable causal inference.
result Performs as well as black-box machine learning methods and better than existing matching methods.
VIBI interprets black-box systems by selecting key features that are both brief and comprehensive.
problem Lack of concise and comprehensive explanations for black-box decision systems.
method VIBI uses the information bottleneck principle to select key features that are maximally compressed and informative.
result VIBI provides more concise and comprehensive explanations compared to existing methods.
Fairwashing occurs when machine learning models are made to appear fair through rationalization.
problem Rationalizing unfair black-box models to appear fair.
method LaundryML uses a regularized rule list enumeration algorithm to find fair rule lists approximating an unfair model.
result It is possible to systematically rationalize decisions from unfair black-box models using model and outcome explanations.
Paper tackles black-box machine teaching with cross-space models, proposing an active teacher model.
problem Teaching a learner with different feature representations and without full observation.
method Proposes an active teacher model that queries the learner to estimate its status and guide faster convergence.
result Active teacher model achieves faster convergence rate than traditional passive learning.
Study efficient black-box attacks on classifiers with limited queries.
problem Attacking machine learning classifiers with query constraints.
method Bayesian optimization to minimize query count.
result Number of queries can be reduced to about 1/10th of random strategy.
New method improves importance sampling for complex proposals.
problem Limited power of traditional importance sampling.
method Black-box importance sampling for any unknown proposal.
result Better and richer proposals improve estimation accuracy.
Develops an axiomatic framework to assess quality of explanation methods for black box machine learning models.
problem Lack of interpretability in black box machine learning models.
method Proposes an axiomatic framework to compare and evaluate the quality of different explanation methods.
result The axiomatic framework is useful for assessing explanation quality and consistent with independent research.
Visual analytics helps interpret machine learning models without sacrificing accuracy.
problem Increasing interpretability of machine learning models decreases predictive power.
method Using visual analytics to interpret black-box machine learning models.
result Visual analytics can help understand model reasoning without reducing predictive quality.
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.
Regularizes black-box models to improve interpretability.
problem Improving interpretability of black-box models without sacrificing accuracy.
method Regularizes a black-box model at training time to connect model explainability, explanation system, and quality metrics.
result Substantial improvement in explanation fidelity and stability across various datasets and explanation systems, with slight accuracy trade-off.
Paper exposes vulnerabilities in interpreting machine learning models using adversarial attacks on PD plots.
problem Vulnerability of permutation-based interpretation methods, particularly PD plots, to adversarial attacks.
method Adversarial framework to manipulate black-box models and produce deceptive PD plots.
result It is possible to hide discriminatory behaviors in machine learning models through interpretation tools like PD plots.
Survey on making machine learning models more understandable.
problem Making predictions from machine learning models transparent and understandable for humans.
method Conducts a survey on explainable supervised machine learning approaches.
result Classifies explainable SML approaches and discusses future directions.
A new AutoML framework uses ADMM to optimize machine learning pipeline configurations.
problem Optimizing machine learning pipelines with mixed integer and continuous variables.
method Leverages ADMM to decompose and solve the optimization problem.
result Significant gains in comparison to other AutoML frameworks.
A method to interpret complex models using a binary tree.
problem Understanding complex machine learning models globally.
method Recursive partitioning of input variable space to maximize contribution differences.
result Improved global interpretation of machine learning models.
A novel method to train networks from each other's adversarial examples to resist black-box attacks.
problem Machine learning models can be fooled by adversarial examples, especially in black-box attacks.
method Simultaneous adversarial training combining two networks to learn from each other's adversarial examples.
result The method improves the networks' resilience to black-box attacks.
Extends batch active learning to non-differentiable models.
problem Efficiently training machine learning models on large, initially unlabelled datasets.
method Black-box batch active learning for regression tasks that relies solely on model predictions.
result Achieves strong performance on regression datasets compared to white-box approaches for deep learning models.
Extracts weighted automata from black box models for sequential data.
problem Global interpretability of black box models for symbolic sequential data.
method Spectral algorithm for extracting weighted automata from black boxes without access to inner representation.
result Approximation of black box models using inferred weighted automata is of high quality.
Paper presents strong black-box attacks on unsupervised clustering models.
problem Adversarial attacks on unsupervised machine learning models.
method Iterative black-box attack on clustering algorithms.
result Adversarial samples can fool clustering algorithms irrespective of the algorithm used.
Paper quantifies how much machine learning models can be explained.
problem Quantifying the explainability of machine learning models.
method Developed a framework to quantify explainability of arbitrary machine learning models.
result Allows for judgment on sufficiency of model explanations.
This paper explores causal analysis in machine learning for better interpretability.
problem The lack of causality in traditional interpretable machine learning models.
method An overview of causal approaches for interpretable machine learning.
result Causal analysis improves the interpretability of machine learning models.
BAR reprograms black-box ML models for transfer learning with scarce data.
problem Transfer learning with limited data and resources.
method Zeroth-order optimization and multi-label mapping techniques to reprogram black-box models.
result BAR outperforms state-of-the-art methods and baseline transfer learning approaches.
Improves interpretability of black-box malware detectors.
problem Fragility of in-vitro evaluations of machine-learning malware detectors.
method Gradient-based approach to identify influential features.
result Enhances interpretability of nonlinear models without sacrificing accuracy.
New method attacks black-box models with fewer queries.
problem Attacking machine learning models without model information.
method Formulated as a real-valued optimization problem, solved with zeroth order optimization.
result Demonstrated to outperform random walk approach on various datasets.
New attack method makes black-box models more vulnerable.
problem Vulnerability of black-box machine learning models to adversarial attacks.
method Decision-based attack that relies solely on model decisions.
result Decision-based attacks are more robust to simple defences than gradient-based attacks.
Influence functions explain black-box model predictions by tracing back to training data.
problem Explain predictions of black-box models.
method Use influence functions from robust statistics to trace predictions back to training data.
result Influence functions provide valuable insights into model behavior and training data.
RuleMatrix visualizes machine learning models for non-experts.
problem Making machine learning models transparent and interpretable for non-expert users.
method Extracts rule-based knowledge from model behavior and presents it in an interactive matrix visualization.
result RuleMatrix helps non-expert users understand and validate machine learning models.
Machine learning impacts computational math, offering new functions approximations.
problem Machine learning's black box nature hinders further progress in computational math.
method Analyzes machine learning's impact on computational math and vice versa.
result Integrating computational math with machine learning can enhance both fields.
Paper finds optimal membership inference strategies for machine learning models.
problem Determining if a sample was part of the training set of a machine learning model.
method Derives optimal strategies for membership inference with assumptions on parameter distribution, showing that black-box attacks are as good as white-box attacks.
result Optimal strategies are not tractable, leading to approximations that outperform existing methods.
Multi-party machine learning leaks global dataset properties even with black-box access.
problem Leakage of global dataset properties in multi-party machine learning.
method Demonstrated leakage of sensitive attribute distributions in pooled data.
result A curious party can infer sensitive attribute distributions in other parties' data with high accuracy.
Paper presents efficient IS for tail risk estimation with machine learning features.
problem Estimating Value at Risk and Conditional Value at Risk with black-box access.
method Efficient Importance Sampling algorithm with self-structuring transformation.
result Asymptotically optimal variance reduction in logarithmic scale.
Defines a new metric to measure importance of predictors in complex machine learning models.
problem Measuring importance of predictors in black box machine learning models.
method Introduces a new metric, GVIM, based on true conditional expectation functions and causal interpretation.
result The GVIM can be represented as a function of Conditional Average Treatment Effect (CATE), providing a causal interpretation.
Unified framework for MCMC and machine learning problems.
problem Intersection of MCMC and machine learning problems.
method Unified framework integrating various MCMC and machine learning techniques.
result Translation and generalization of theory and methods.
Paper tackles constrained robust optimization without gradients.
problem Constrained robust (min-max) optimization in a black-box setting.
method Integrates ZO gradient estimator with alternating projected SGD-ascent method.
result Proposed ZO-Min-Max framework converges sub-linearly under mild conditions.
Paper presents mdfa to identify victims of discrimination in black box classifiers.
problem Identifying victims of discrimination in black box classifiers.
method Reduces discrimination measurement to matching distributions and sensitive attribute coincidence prediction.
result Identifies African-American individuals at high risk of violent recidivism.
Proposes a new method to explain complex machine learning models.
problem Lack of joint feature effects in current explanation algorithms.
method Axiomatization of the Banzhaf index for capturing feature subsets.
result The generalized Banzhaf index optimally approximates black-box models.
New estimator optimizes black-box model errors in semiparametric estimation.
problem How nuisance estimation errors affect low-dimensional target parameters in semiparametric models.
method Proposed a new estimator achieving a sharper rate of convergence.
result The first-order stochastic error of nuisance estimation can be eliminated.
Deep learning reduces complex data to simpler predictors.
problem High-dimensional data reduction in input-output models.
method Hierarchical layers of latent features for constructing predictors.
result Deep learning is a black-box method for high-dimensional function estimation.
New method for interpreting complex ML models.
problem Interpreting complex black-box ML models.
method Functional decomposition of black-box predictions into simpler subfunctions.
result Main effects provide insights into feature contributions and interactions.
SurvBeX explains ML survival models using Beran estimator.
problem Interpreting predictions of machine learning survival models.
method Uses modified Beran estimator as surrogate model to compute feature impacts.
result SurvBeX minimizes mean distance between black-box and surrogate model survival functions.
This paper reviews zeroth-order optimization in signal processing and machine learning.
problem Optimization problems without gradient information.
method Iterative steps: gradient estimation, descent direction computation, solution update.
result Demonstrates applications in robustness evaluation and black-box model explanations.
Method for explaining machine learning survival models using counterfactuals.
problem Tackles the challenge of explaining survival models in machine learning.
method Introduces a condition based on the difference of mean times to event for counterfactual explanation. Reduces the problem to a convex optimization problem for Cox models and applies Particle Swarm Optimization for other models.
result Demonstrates the effectiveness of the proposed method through numerical experiments.
VINE visualizes statistical interactions in complex models.
problem Lack of utilities for regional explanations in black box models.
method VINE algorithm to extract and visualize statistical interaction effects.
result VINE provides a novel evaluation metric for visualizations.
HDMR provides insights into machine learning models, aiding in both prediction and explanation.
problem Understanding and interpreting complex machine learning models.
method High Dimensional Model Representation (HDMR) and its applications in machine learning.
result HDMR offers a glass box approach to machine learning models, enhancing both prediction and explanation.
Paper explores how black box models can deviate from average performance.
problem Understanding and interpreting predictions from sophisticated black box models.
method Two general approaches to provide interpretable descriptions of black box classification model performance.
result Identifies regions where black box models deviate significantly from their average performance.
Bayesian optimisation tackles expensive black-box functions with constraints.
problem Optimizing constrained black-box functions in machine learning and simulation.
method Proposes a new Knowledge Gradient acquisition function for constrained Bayesian optimisation.
result Demonstrates superior performance over four state-of-the-art constrained Bayesian optimisation algorithms.