A new confidence scoring method using whitebox models and linear probes.
problem Improving confidence scores for deep neural networks.
method Two-model paradigm with a base model and a meta-model using linear classifier probes.
result The method outperforms baselines in filtering tasks.
New method tests tree models without causing computational pressure.
problem Incompatibility of adversarial robustness testing with tree ensembles.
method Smooths tree ensembles with sigmoid functions and uses gradient descent.
result Successfully reveals adversarial vulnerability of tree ensemble models.
Random Gaussian noise and pixel discretization improve image classifier robustness.
problem Whitebox adversarial attacks decrease classifier accuracy.
method Inject random Gaussian noise, discretize pixels, and use any classifier.
result Reduces KL divergence and lower bound on classifier accuracy.
A model disentangles features to detect adversarial inputs.
problem Detecting and defending against adversarial attacks on neural networks.
method Proposes a minimax game formulation using variational autoencoders to separate robust and vulnerable features.
result Adversarial inputs cannot bypass the detector without semantic change, indicating successful detection.
AdvGAN generates high-quality adversarial examples using GANs.
problem Vulnerability of deep neural networks to adversarial examples.
method Generative adversarial networks (GANs) to learn and approximate input distributions.
result AdvGAN achieves high attack success rates on various models.
Noise added to deep Q-networks reduces adversarial attacks.
problem Vulnerability of deep reinforcement learning to policy manipulation attacks.
method Addition of noise to the parameter space of deep reinforcement learners during training.
result Noise reduces the transferability of adversarial examples.
New method defends deep learning models against adversarial attacks.
problem Robustness of deep learning models against adversarial attacks is compromised.
method Adversarial samples are relaxed onto the target class distribution manifold using MALA.
result MALADE outperformed state-of-the-art methods in various attacks.
This work generates diverse adversarial attacks for different domains using latent variable perturbation.
problem Adversarial attacks on deep neural networks are limited to a single perturbation.
method Frame adversarial attacks as learning a distribution of perturbations, enabling generation of diverse attacks.
result Framework generates competitive or superior adversarial attacks across diverse domains (images, text, graphs).
Adversarial perturbations fool deepfake detectors with high accuracy.
problem Improving deepfake detection accuracy against adversarial attacks.
method Used adversarial perturbations and two defenses: Lipschitz regularization and Deep Image Prior (DIP).
result Deepfake detectors achieved 27% accuracy on perturbed images, compared to 95% on unperturbed.
Proposes a query-efficient blackbox attack method.
problem Adversarial attacks on machine learning models, especially deep neural networks.
method QEBA: Query-Efficient Boundary-based blackbox Attack, using only final prediction labels.
result QEBA achieves 100% attack success rate with fewer queries and lower perturbation.
WAFFLe anonymizes federated learning weights to protect data privacy and fairness.
problem Federated learning exposes local models to attacks and underfits heterogeneous clients.
method Combines Indian Buffet Process with shared weight factors.
result Significant improvement in local test performance and fairness.
A new method defends against adversarial examples using minimax optimization.
problem Adversarial examples can fool state-of-the-art classifiers.
method Formulated as a two-player game, proposed minimax optimization algorithm.
result Numerical minimax defense is more robust than non-minimax defenses.
New auditors assess f-DP privacy with adaptive sampling, avoiding large sample sizes.
problem Empirical auditing of f-DP privacy with adaptive sampling. method Shift focus to f-DP, develop adaptive auditors for whitebox and blackbox settings. result Adaptive auditors detect f-DP violations across the privacy spectrum with statistical guarantees. Lower bounds on query complexity for reconstructing private learner's training data.
problem Query complexity of reconstructing private learner's training data.
method Minimax analysis, Rényi DP, Metric DP framework.
result First known lower bounds on adversary's query complexity for various DP learners.
This paper improves Prior Networks for uncertainty and adversarial robustness.
problem Improving uncertainty and adversarial robustness in classification models.
method Introduces reverse KL-divergence for training Prior Networks and proposes a generalized adversarial training method.
result Prior Networks trained with reverse KL-divergence show improved out-of-distribution detection and are harder to evade in adversarial attacks.
Prior Networks improve adversarial attack detection by estimating uncertainty.
problem Detecting adversarial attacks in AI systems, especially those designed to evade detection methods.
method Prior Networks estimate predictive uncertainty, outperforming Monte-Carlo dropout ensembles.
result Prior Networks significantly outperform baseline approaches in detecting adversarial attacks.
Study develops time-continuous models and probabilistic descriptions for agent-based economic market models.
problem Formulating and describing agent-based economic market models in a time-continuous and probabilistic manner.
method Derived time-continuous formulations, discussed impact of time-scaling, proved stability, presented probabilistic descriptions using kinetic theory.
result Time-continuous formulations and probabilistic descriptions for agent-based economic market models.
Hybrid model combines interpretable and black-box models for better transparency and performance.
problem Balancing interpretability and predictive performance in machine learning models.
method Proposes a Hybrid Predictive Model (HPM) integrating an interpretable model with a black-box model, using principled objective functions and customized training algorithms.
result Hybrid models achieve an efficient trade-off between transparency and predictive performance.
The paper introduces BCART models for aggregate claim amount, improving frequency-severity and joint modeling.
problem Modeling aggregate claim amount with frequency-severity and joint dependencies.
method Developed three types of BCART models: frequency-severity, sequential, and joint models. Used various distributions for claim severity data.
result Weibull distribution outperforms gamma and lognormal for right-skewed, heavy-tailed claim severity data.
Boosts generative models by combining multiple meta-models.
problem Challenges in creating a single generative model that accurately represents complex data.
method Cascades multiple meta-models (like RBM and VAE) to create a stronger generative model.
result Derives a decomposable variational lower bound for training and evaluating the boosted model.
The paper uses model-based trees to create interpretable surrogate models for complex machine learning models.
problem Interpreting complex machine learning models.
method Using model-based trees to partition feature space and create interpretable models.
result Model-based trees generate optimal surrogate models that balance interpretability and performance.
Study on limits of community detection in various network models.
problem Limits of community detection in network models.
method Analysis of several network models including Stochastic Block Model, Exponential Random Graph Model, Latent Space Model, Directed Preferential Attachment Model, and Directed Small-world Model.
result Information-theoretic limits for recovery of node labels in network models.
Gauge Flow Models use a learnable Gauge Field in Generative Flow Models.
problem Improving generative model performance.
method Integrates a learnable Gauge Field into Flow ODEs.
result Gauge Flow Models outperform traditional Flow Models in Flow Matching experiments.
The study examines how model predictions hold up under model extensions.
problem Model predictions may not be robust under model extensions, limiting their applicability.
method The study uses causal ordering to assess robustness of qualitative model predictions and characterizes model extensions that preserve predictions.
result Conditions and techniques are provided to assess robustness of model predictions under model extensions.
MALC combines interpretable linear models with black-box models for better predictions and transparency.
problem Combining interpretability with black-box models for better predictions.
method Formulates MALC as a convex optimization problem and uses accelerated proximal gradient method for training.
result MALC provides an efficient frontier balancing prediction accuracy and transparency.
Alternative approach to model selection using transformation analysis.
problem Over-simplistic models lead to erroneous interpretations.
method Step-wise complexity reduction to identify simpler, better-interpretable models.
result Transformation models improve model fit and interpretability.
Revises Bayesian model averaging for foundation models.
problem Ensemble pre-trained and lightly-finetuned foundation models for improved classification performance.
method Introduces trainable linear classifiers and computationally cheaper model averaging scheme (OMA).
result Ensembled models can better predict on various datasets.
Paper introduces symmetric divergence link models for probability distributions.
problem Symmetric divergence measures for probability distributions.
method Two general classes of link models: one for survival functions and another for cumulative probability distribution functions.
result Advantages of symmetric divergence measures over asymmetric measures for model averaging and feature assessment.
The paper tests stock return models and uses LSTM to predict stock returns.
problem Validating stock return models and predicting stock returns.
method Used Fama-French three-factor, four-factor, and five-factor models; also used LSTM model.
result Fama-French five-factor model shows better validity for stock returns.
New method to handle credit portfolio model uncertainties.
problem Model risk in credit portfolio models.
method Demonstrates comprehensive yet easy-to-implement approach to uncertainty in model parameters.
result Comprehensive method to deal with model uncertainties.
A new neural network model predicts multi-symbol tokens over multiple scales.
problem Language modeling with improved flexibility and performance.
method A learned dictionary of multi-symbol tokens using BPE compression.
result The model outperforms LSTM on language modeling tasks, especially for smaller models.
Researchers review challenges in interpreting additive models, especially neural additive models.
problem Challenges in interpreting additive models, particularly neural additive models.
method Review of generalized additive models and discussion of nonidentifiability.
result Challenges in claiming interpretability or suitability for safety-critical applications of additive models.
Distill-and-Compare audits black-box models by training transparent models to mimic them.
problem Auditing proprietary, opaque black-box risk scoring models.
method Model distillation and comparison of transparent student models to black-box models.
result Identifies missing features in black-box models, improving transparency.
Proposes a decision-theoretic approach for enhancing model interpretability in Bayesian frameworks.
problem Challenges the traditional approach of restricting model structure for interpretability in Bayesian frameworks.
method Introduces an interpretability utility function and a two-step method involving a reference model and a proxy model.
result Demonstrates that the proposed method generates more accurate models with the same level of interpretability.
Novel hybrid modeling combines ML and physics for real-time diagnosis.
problem Real-time diagnosis of complex systems.
method Combines machine learning and physics-based models to create reduced-order models.
result Generated models are two orders of magnitude simpler, improving efficiency.
CRS model improves ranking data modeling with theoretical guarantees.
problem Lack of rich, multimodal models for ranking data.
method Contextual Repeated Selection (CRS) model for multimodal ranking data.
result CRS model significantly outperforms existing methods in various ranking contexts.
Sigma models linked to Gross-Neveu models via quiver varieties.
problem Understanding the relationship between sigma models and Gross-Neveu models.
method Exploring the mathematical correspondence between sigma models and Gross-Neveu models, including their geometric and trigonometric/elliptic deformations.
result Sigma models are mathematically equivalent to Gross-Neveu models under certain conditions.
Matryoshka hides secret models in a carrier model, achieving high capacity and robustness.
problem Stealing functionality of private ML data by hiding models in a carrier model.
method Parameter sharing approach exploiting the learning capacity of the carrier model.
result Hides a 26x larger secret model or 8 secret models in the carrier model.
New Cartan model for equivariant cohomology developed.
problem Developing a new framework for equivariant cohomology.
method Introducing a new operator dC and constructing a Cartan model. result Relations between new BRST and Weil models established.
Eigen-stratified models reduce model size and improve performance.
problem Large model size in Laplacian-regularized stratified models.
method Formulate eigen-stratified models with linear combinations of bottom eigenvectors of the graph Laplacian.
result Significant reduction in model size with eigen-stratified models.
Semi-parametric models improve robot dynamics modeling accuracy.
problem Improving inverse dynamics model accuracy in robotics.
method Comparison of semi-parametric Gaussian process regression and a novel model-based neural network.
result Semi-parametric Gaussian process regression yields the most accurate models.
A novel kernel approach for model selection in simulator-based models.
problem Model selection for simulator-based statistical models with limited prior knowledge.
method Iteratively updates model weights and parameters using Bayes' rule and kernel recursive ABC algorithm.
result Demonstrates effectiveness on dynamical systems in ecology and epidemiology.
This work develops scalable model selection methods with fast update and selection.
problem Efficient model selection for large pools of candidate models.
method Isolated model embedding, which supports asymptotically fast update and selection.
result Standardized Embedder achieves competitive model selection performances.
Seq2Seq models speed up epidemic model predictions.
problem Complex epidemic models are computationally expensive.
method Used deep seq2seq models as surrogates for complex models.
result Surrogates predict scenarios up to several thousand times faster.
Copulas outperform marginal models in multivariate risk forecasting, reducing model risk by narrowing down the set of models.
problem Model risk in multivariate risk forecasting, especially during crises.
method Comprehensive empirical study comparing Copula-GARCH models with fixed marginals, copulas, or neither.
result Model risk is almost entirely due to copula choice, not marginal models.
Paper proposes BMPO to optimize policies using bidirectional models.
problem Model-based reinforcement learning's reliance on forward model accuracy.
method Develops BMPO using both forward and backward models for policy optimization.
result BMPO outperforms state-of-the-art methods in sample efficiency and asymptotic performance.
Model extraction simplifies complex models for easier interpretation.
problem Interpreting complex machine learning models for consequential decisions.
method Approximating complex models with interpretable models to preserve statistical properties.
result Model extraction effectively interprets random forests and neural nets.
This paper distills a complex travel mode choice model into simpler, interpretable models.
problem Lack of interpretability in complex machine learning models for travel behavior.
method Model distillation combined with market segmentation.
result Generated interpretable models that closely match the predictions of the original complex model.