Search Results for author:
Found 55 papers, 34 papers with code
One-shot Distributed Ridge Regression in High Dimensions
To scale up data analysis, distributed and parallel computing approaches are increasingly needed.
Uncertainty in Language Models: Assessment through Rank-Calibration
Language Models (LMs) have shown promising performance in natural language generation.
Inference in Randomized Least Squares and PCA via Normality of Quadratic Forms
To our knowledge, no comparable methods are available for SSE and for SRHT in PCA.
JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models
To address these challenges, we introduce JailbreakBench, an open-sourced benchmark with the following components: (1) an evolving repository of state-of-the-art adversarial prompts, which we refer to as jailbreak artifacts; (2) a jailbreaking dataset comprising 100 behaviors -- both original and sourced from prior work -- which align with OpenAI's usage policies; (3) a standardized evaluation framework that includes a clearly defined threat model, system prompts, chat templates, and scoring functions; and (4) a leaderboard that tracks the performance of attacks and defenses for various LLMs.
Minimax Optimal Fair Classification with Bounded Demographic Disparity
Mitigating the disparate impact of statistical machine learning methods is crucial for ensuring fairness.
Bayes-Optimal Fair Classification with Linear Disparity Constraints via Pre-, In-, and Post-processing
To address this, we develop methods for Bayes-optimal fair classification, aiming to minimize classification error subject to given group fairness constraints.
SymmPI: Predictive Inference for Data with Group Symmetries
Methods for predictive inference have been developed under a variety of assumptions, often -- for instance, in standard conformal prediction -- relying on the invariance of the distribution of the data under special groups of transformations such as permutation groups.
PAC Prediction Sets Under Label Shift
We propose a novel algorithm for constructing prediction sets with PAC guarantees in the label shift setting.
Jailbreaking Black Box Large Language Models in Twenty Queries
PAIR -- which is inspired by social engineering attacks -- uses an attacker LLM to automatically generate jailbreaks for a separate targeted LLM without human intervention.
A Theory of Non-Linear Feature Learning with One Gradient Step in Two-Layer Neural Networks
However, with a constant gradient descent step size, this spike only carries information from the linear component of the target function and therefore learning non-linear components is impossible.
Statistical Estimation Under Distribution Shift: Wasserstein Perturbations and Minimax Theory
Under a squared loss for mean estimation and prediction error in linear regression, we find the exact minimax risk, a least favorable perturbation, and show that the sample mean and least squares estimators are respectively optimal.
Efficient and Multiply Robust Risk Estimation under General Forms of Dataset Shift
Despite extensive literature on dataset shift, limited works address how to efficiently use the auxiliary populations to improve the accuracy of risk evaluation for a given machine learning task in the target population.
Optimal Heterogeneous Collaborative Linear Regression and Contextual Bandits
MOLAR improves the dependence of the estimation error on the data dimension, compared to independent least squares estimates.
Sharp-SSL: Selective high-dimensional axis-aligned random projections for semi-supervised learning
We propose a new method for high-dimensional semi-supervised learning problems based on the careful aggregation of the results of a low-dimensional procedure applied to many axis-aligned random projections of the data.
Demystifying Disagreement-on-the-Line in High Dimensions
Evaluating the performance of machine learning models under distribution shift is challenging, especially when we only have unlabeled data from the shifted (target) domain, along with labeled data from the original (source) domain.
Conformal Frequency Estimation using Discrete Sketched Data with Coverage for Distinct Queries
This paper develops conformal inference methods to construct a confidence interval for the frequency of a queried object in a very large discrete data set, based on a sketch with a lower memory footprint.
PAC Prediction Sets for Meta-Learning
Uncertainty quantification is a key component of machine learning models targeted at safety-critical systems such as in healthcare or autonomous vehicles.
Pursuit of a Discriminative Representation for Multiple Subspaces via Sequential Games
We consider the problem of learning discriminative representations for data in a high-dimensional space with distribution supported on or around multiple low-dimensional linear subspaces.
Unified Fourier-based Kernel and Nonlinearity Design for Equivariant Networks on Homogeneous Spaces
We present a unified derivation of kernels via the Fourier domain by leveraging the sparsity of Fourier coefficients of the lifted feature fields.
Memory Classifiers: Two-stage Classification for Robustness in Machine Learning
We propose a new method for classification which can improve robustness to distribution shifts, by combining expert knowledge about the ``high-level" structure of the data with standard classifiers.
Collaborative Learning of Discrete Distributions under Heterogeneity and Communication Constraints
In modern machine learning, users often have to collaborate to learn the distribution of the data.
PAC-Wrap: Semi-Supervised PAC Anomaly Detection
Anomaly detection is essential for preventing hazardous outcomes for safety-critical applications like autonomous driving.
Fair Bayes-Optimal Classifiers Under Predictive Parity
This paper considers predictive parity, which requires equalizing the probability of success given a positive prediction among different protected groups.
SE(3)-Equivariant Attention Networks for Shape Reconstruction in Function Space
In contrast to previous shape reconstruction methods that align the input to a regular grid, we operate directly on the irregular point cloud.
Prediction Sets Adaptive to Unknown Covariate Shift
Predicting sets of outcomes -- instead of unique outcomes -- is a promising solution to uncertainty quantification in statistical learning.
T-Cal: An optimal test for the calibration of predictive models
We find that detecting mis-calibration is only possible when the conditional probabilities of the classes are sufficiently smooth functions of the predictions.
Exploring with Sticky Mittens: Reinforcement Learning with Expert Interventions via Option Templates
We formulate expert intervention as allowing the agent to execute option templates before learning an implementation.
Bayes-Optimal Classifiers under Group Fairness
Machine learning algorithms are becoming integrated into more and more high-stakes decision-making processes, such as in social welfare issues.
iDECODe: In-distribution Equivariance for Conformal Out-of-distribution Detection
We propose the new method iDECODe, leveraging in-distribution equivariance for conformal OOD detection.
Learning Augmentation Distributions using Transformed Risk Minimization
We propose a new \emph{Transformed Risk Minimization} (TRM) framework as an extension of classical risk minimization.
Solon: Communication-efficient Byzantine-resilient Distributed Training via Redundant Gradients
There has been a growing need to provide Byzantine-resilience in distributed model training.
Comparing Classes of Estimators: When does Gradient Descent Beat Ridge Regression in Linear Models?
Methods for learning from data depend on various types of tuning parameters, such as penalization strength or step size.
PAC Prediction Sets Under Covariate Shift
Our approach focuses on the setting where there is a covariate shift from the source distribution (where we have labeled training examples) to the target distribution (for which we want to quantify uncertainty).
Understanding Generalization in Adversarial Training via the Bias-Variance Decomposition
To investigate this gap, we decompose the test risk into its bias and variance components and study their behavior as a function of adversarial training perturbation radii ($\varepsilon$).
Optimal Iterative Sketching Methods with the Subsampled Randomized Hadamard Transform
These show that the convergence rate for Haar and randomized Hadamard matrices are identical, and asymptotically improve upon Gaussian random projections.
Sparse sketches with small inversion bias
For a tall $n\times d$ matrix $A$ and a random $m\times n$ sketching matrix $S$, the sketched estimate of the inverse covariance matrix $(A^\top A)^{-1}$ is typically biased: $E[(\tilde A^\top\tilde A)^{-1}]\ne(A^\top A)^{-1}$, where $\tilde A=SA$.
What causes the test error? Going beyond bias-variance via ANOVA
This leads to discovering the unimodality of variance as a function of the level of parametrization, and to decomposing the variance into that arising from label noise, initialization, and randomness in the training data to understand the sources of the error.
DeltaGrad: Rapid retraining of machine learning models
Machine learning models are not static and may need to be retrained on slightly changed datasets, for instance, with the addition or deletion of a set of data points.
Provable tradeoffs in adversarially robust classification
It is well known that machine learning methods can be vulnerable to adversarially-chosen perturbations of their inputs.
The Implicit Regularization of Stochastic Gradient Flow for Least Squares
We study the implicit regularization of mini-batch stochastic gradient descent, when applied to the fundamental problem of least squares regression.
Optimal Iterative Sketching with the Subsampled Randomized Hadamard Transform
These show that the convergence rate for Haar and randomized Hadamard matrices are identical, and asymptotically improve upon Gaussian random projections.
Implicit Regularization and Convergence for Weight Normalization
For certain stepsizes of g and w , we show that they can converge close to the minimum norm solution.
Ridge Regression: Structure, Cross-Validation, and Sketching
(2) how to correctly use cross-validation to choose the regularization parameter?
A Group-Theoretic Framework for Data Augmentation
Data augmentation is a widely used trick when training deep neural networks: in addition to the original data, properly transformed data are also added to the training set.
WONDER: Weighted one-shot distributed ridge regression in high dimensions
Here we study a fundamental and highly important problem in this area: How to do ridge regression in a distributed computing environment?
Asymptotics for Sketching in Least Squares Regression
We consider a least squares regression problem where the data has been generated from a linear model, and we are interested to learn the unknown regression parameters.
Distributed linear regression by averaging
Here we study the performance loss in estimation, test error, and confidence interval length in high dimensions, where the number of parameters is comparable to the training data size.
Robust Inference Under Heteroskedasticity via the Hadamard Estimator
In this paper, we propose methods that are robust to large and unequal noise in different observational units (i. e., heteroskedasticity) for statistical inference in linear regression.
Statistics Theory Methodology Statistics Theory
Flexible Multiple Testing with the FACT Algorithm
Modern high-throughput science often leads to multiple testing problems: researchers test many hypotheses, wishing to find the significant discoveries.
Methodology
Deterministic parallel analysis
This paper presents a deterministic version of PA (DPA), which is faster and more reproducible than PA. We show that DPA selects large factors and does not select small factors just like [Dobriban, 2017] shows for PA.
Methodology
Permutation methods for factor analysis and PCA
In this paper, we show that the parallel analysis permutation method consistently selects the large components in certain high-dimensional factor models.
Statistics Theory Methodology Statistics Theory
$e$PCA: High Dimensional Exponential Family PCA
We develop $e$PCA (exponential family PCA), a new methodology for PCA on exponential family distributions.
Methodology
High-Dimensional Asymptotics of Prediction: Ridge Regression and Classification
We provide a unified analysis of the predictive risk of ridge regression and regularized discriminant analysis in a dense random effects model.
Efficient Computation of Limit Spectra of Sample Covariance Matrices
Asymptotically, as $n, p \to \infty$ with $p/n \to \gamma$, there is a deterministic mapping from the population spectral distribution (PSD) to the empirical spectral distribution (ESD) of the eigenvalues.
Numerical Analysis Probability
Optimal Multiple Testing Under a Gaussian Prior on the Effect Sizes
For a Gaussian prior on effect sizes, we show that finding the optimal weights is a non-convex problem.
Methodology
