Search Results for author:
Found 33 papers, 13 papers with code
Algorithmic Arbitrariness in Content Moderation
Our findings also contribute to content moderation and intermediary liability laws being discussed and passed in many countries, such as the Digital Services Act in the European Union, the Online Safety Act in the United Kingdom, and the Fake News Bill in Brazil.
Interpreting CLIP with Sparse Linear Concept Embeddings (SpLiCE)
CLIP embeddings have demonstrated remarkable performance across a wide range of computer vision tasks.
Multi-Group Fairness Evaluation via Conditional Value-at-Risk Testing
Here, the sample complexity for estimating the worst-case performance gap across groups (e. g., the largest difference in error rates) increases exponentially with the number of group-denoting sensitive attributes.
Differentially Private Secure Multiplication: Hiding Information in the Rubble of Noise
We consider the problem of private distributed multi-party multiplication.
Fair Machine Unlearning: Data Removal while Mitigating Disparities
In this work, we demonstrate that most efficient unlearning methods cannot accommodate popular fairness interventions, and we propose the first fair machine unlearning method that can efficiently unlearn data instances from a fair objective.
Arbitrariness Lies Beyond the Fairness-Accuracy Frontier
Machine learning tasks may admit multiple competing models that achieve similar performance yet produce conflicting outputs for individual samples -- a phenomenon known as predictive multiplicity.
Gaussian Max-Value Entropy Search for Multi-Agent Bayesian Optimization
We focus on Entropy Search (ES), a sample-efficient BO algorithm that selects queries to maximize the mutual information about the maximum of the black-box function.
Arbitrary Decisions are a Hidden Cost of Differentially Private Training
We demonstrate that such randomization incurs predictive multiplicity: for a given input example, the output predicted by equally-private models depends on the randomness used in training.
Aleatoric and Epistemic Discrimination: Fundamental Limits of Fairness Interventions
We categorize sources of discrimination in the ML pipeline into two classes: aleatoric discrimination, which is inherent in the data distribution, and epistemic discrimination, which is due to decisions made during model development.
Automated Segmentation and Recurrence Risk Prediction of Surgically Resected Lung Tumors with Adaptive Convolutional Neural Networks
Then, the identified tumors are passed to a second CNN for recurrence risk prediction.
The Saddle-Point Accountant for Differential Privacy
SPA approximates privacy guarantees for the composition of DP mechanisms in an accurate and fast manner.
Bottlenecks CLUB: Unifying Information-Theoretic Trade-offs Among Complexity, Leakage, and Utility
In this work, we propose a general family of optimization problems, termed as complexity-leakage-utility bottleneck (CLUB) model, which (i) provides a unified theoretical framework that generalizes most of the state-of-the-art literature for the information-theoretic privacy models, (ii) establishes a new interpretation of the popular generative and discriminative models, (iii) constructs new insights to the generative compression models, and (iv) can be used in the fair generative models.
Cactus Mechanisms: Optimal Differential Privacy Mechanisms in the Large-Composition Regime
Since the optimization problem is infinite dimensional, it cannot be solved directly; nevertheless, we quantize the problem to derive near-optimal additive mechanisms that we call "cactus mechanisms" due to their shape.
Beyond Adult and COMPAS: Fairness in Multi-Class Prediction
We consider the problem of producing fair probabilistic classifiers for multi-class classification tasks.
Polynomial Approximations of Conditional Expectations in Scalar Gaussian Channels
We consider a channel $Y=X+N$ where $X$ is a random variable satisfying $\mathbb{E}[|X|]<\infty$ and $N$ is an independent standard normal random variable.
Information Theory Information Theory Probability
Generalization Bounds for Noisy Iterative Algorithms Using Properties of Additive Noise Channels
In this paper, we analyze the generalization of models trained by noisy iterative algorithms.
Local Differential Privacy Is Equivalent to Contraction of $E_γ$-Divergence
We investigate the local differential privacy (LDP) guarantees of a randomized privacy mechanism via its contraction properties.
Contraction of $E_γ$-Divergence and Its Applications to Privacy
First, it implies that local differential privacy can be equivalently expressed in terms of the contraction of $E_\gamma$-divergence.
Three Variants of Differential Privacy: Lossless Conversion and Applications
In the first part, we develop a machinery for optimally relating approximate DP to RDP based on the joint range of two $f$-divergences that underlie the approximate DP and RDP.
CPR: Classifier-Projection Regularization for Continual Learning
Inspired by both recent results on neural networks with wide local minima and information theory, CPR adds an additional regularization term that maximizes the entropy of a classifier's output probability.
To Split or Not to Split: The Impact of Disparate Treatment in Classification
To evaluate the effect of disparate treatment, we compare the performance of split classifiers (i. e., classifiers trained and deployed separately on each group) with group-blind classifiers (i. e., classifiers which do not use a sensitive attribute).
Privacy-Preserving Image Sharing via Sparsifying Layers on Convolutional Groups
We propose a practical framework to address the problem of privacy-aware image sharing in large-scale setups.
Privacy Amplification of Iterative Algorithms via Contraction Coefficients
We investigate the framework of privacy amplification by iteration, recently proposed by Feldman et al., from an information-theoretic lens.
A Better Bound Gives a Hundred Rounds: Enhanced Privacy Guarantees via $f$-Divergences
We derive the optimal differential privacy (DP) parameters of a mechanism that satisfies a given level of R\'enyi differential privacy (RDP).
Correspondence Analysis Using Neural Networks
Correspondence analysis (CA) is a multivariate statistical tool used to visualize and interpret data dependencies.
Repairing without Retraining: Avoiding Disparate Impact with Counterfactual Distributions
When the performance of a machine learning model varies over groups defined by sensitive attributes (e. g., gender or ethnicity), the performance disparity can be expressed in terms of the probability distributions of the input and output variables over each group.
Correspondence Analysis of Government Expenditure Patterns
We analyze expenditure patterns of discretionary funds by Brazilian congress members.
Generalizing Correspondence Analysis for Applications in Machine Learning
In this paper, we provide a novel interpretation of CA in terms of an information-theoretic quantity called the principal inertia components.
Generalizing Bottleneck Problems
Given a pair of random variables $(X, Y)\sim P_{XY}$ and two convex functions $f_1$ and $f_2$, we introduce two bottleneck functionals as the lower and upper boundaries of the two-dimensional convex set that consists of the pairs $\left(I_{f_1}(W; X), I_{f_2}(W; Y)\right)$, where $I_f$ denotes $f$-information and $W$ varies over the set of all discrete random variables satisfying the Markov condition $W \to X \to Y$.
On the Direction of Discrimination: An Information-Theoretic Analysis of Disparate Impact in Machine Learning
In the context of machine learning, disparate impact refers to a form of systematic discrimination whereby the output distribution of a model depends on the value of a sensitive attribute (e. g., race or gender).
Privacy with Estimation Guarantees
Here, an analyst is allowed to reconstruct (in a mean-squared error sense) certain functions of the data (utility), while other private functions should not be reconstructed with distortion below a certain threshold (privacy).
Optimized Data Pre-Processing for Discrimination Prevention
Non-discrimination is a recognized objective in algorithmic decision making.
