Alert button
Picture for Edgar Dobriban

Edgar Dobriban

Alert button

PAC Prediction Sets Under Label Shift

Oct 19, 2023
Wenwen Si, Sangdon Park, Insup Lee, Edgar Dobriban, Osbert Bastani

Prediction sets capture uncertainty by predicting sets of labels rather than individual labels, enabling downstream decisions to conservatively account for all plausible outcomes. Conformal inference algorithms construct prediction sets guaranteed to contain the true label with high probability. These guarantees fail to hold in the face of distribution shift, which is precisely when reliable uncertainty quantification can be most useful. We propose a novel algorithm for constructing prediction sets with PAC guarantees in the label shift setting. This method estimates the predicted probabilities of the classes in a target domain, as well as the confusion matrix, then propagates uncertainty in these estimates through a Gaussian elimination algorithm to compute confidence intervals for importance weights. Finally, it uses these intervals to construct prediction sets. We evaluate our approach on five datasets: the CIFAR-10, ChestX-Ray and Entity-13 image datasets, the tabular CDC Heart dataset, and the AGNews text dataset. Our algorithm satisfies the PAC guarantee while producing smaller, more informative, prediction sets compared to several baselines.

Viaarxiv icon

Jailbreaking Black Box Large Language Models in Twenty Queries

Oct 13, 2023
Patrick Chao, Alexander Robey, Edgar Dobriban, Hamed Hassani, George J. Pappas, Eric Wong

Figure 1 for Jailbreaking Black Box Large Language Models in Twenty Queries
Figure 2 for Jailbreaking Black Box Large Language Models in Twenty Queries
Figure 3 for Jailbreaking Black Box Large Language Models in Twenty Queries
Figure 4 for Jailbreaking Black Box Large Language Models in Twenty Queries

There is growing interest in ensuring that large language models (LLMs) align with human values. However, the alignment of such models is vulnerable to adversarial jailbreaks, which coax LLMs into overriding their safety guardrails. The identification of these vulnerabilities is therefore instrumental in understanding inherent weaknesses and preventing future misuse. To this end, we propose Prompt Automatic Iterative Refinement (PAIR), an algorithm that generates semantic jailbreaks with only black-box access to an LLM. PAIR -- which is inspired by social engineering attacks -- uses an attacker LLM to automatically generate jailbreaks for a separate targeted LLM without human intervention. In this way, the attacker LLM iteratively queries the target LLM to update and refine a candidate jailbreak. Empirically, PAIR often requires fewer than twenty queries to produce a jailbreak, which is orders of magnitude more efficient than existing algorithms. PAIR also achieves competitive jailbreaking success rates and transferability on open and closed-source LLMs, including GPT-3.5/4, Vicuna, and PaLM-2.

* 21 pages, 10 figures 
Viaarxiv icon

A Theory of Non-Linear Feature Learning with One Gradient Step in Two-Layer Neural Networks

Oct 11, 2023
Behrad Moniri, Donghwan Lee, Hamed Hassani, Edgar Dobriban

Figure 1 for A Theory of Non-Linear Feature Learning with One Gradient Step in Two-Layer Neural Networks
Figure 2 for A Theory of Non-Linear Feature Learning with One Gradient Step in Two-Layer Neural Networks
Figure 3 for A Theory of Non-Linear Feature Learning with One Gradient Step in Two-Layer Neural Networks

Feature learning is thought to be one of the fundamental reasons for the success of deep neural networks. It is rigorously known that in two-layer fully-connected neural networks under certain conditions, one step of gradient descent on the first layer followed by ridge regression on the second layer can lead to feature learning; characterized by the appearance of a separated rank-one component -- spike -- in the spectrum of the feature matrix. 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. We show that with a learning rate that grows with the sample size, such training in fact introduces multiple rank-one components, each corresponding to a specific polynomial feature. We further prove that the limiting large-dimensional and large sample training and test errors of the updated neural networks are fully characterized by these spikes. By precisely analyzing the improvement in the loss, we demonstrate that these non-linear features can enhance learning.

Viaarxiv icon

Statistical Estimation Under Distribution Shift: Wasserstein Perturbations and Minimax Theory

Aug 03, 2023
Patrick Chao, Edgar Dobriban

Figure 1 for Statistical Estimation Under Distribution Shift: Wasserstein Perturbations and Minimax Theory
Figure 2 for Statistical Estimation Under Distribution Shift: Wasserstein Perturbations and Minimax Theory
Figure 3 for Statistical Estimation Under Distribution Shift: Wasserstein Perturbations and Minimax Theory
Figure 4 for Statistical Estimation Under Distribution Shift: Wasserstein Perturbations and Minimax Theory

Distribution shifts are a serious concern in modern statistical learning as they can systematically change the properties of the data away from the truth. We focus on Wasserstein distribution shifts, where every data point may undergo a slight perturbation, as opposed to the Huber contamination model where a fraction of observations are outliers. We formulate and study shifts beyond independent perturbations, exploring Joint Distribution Shifts, where the per-observation perturbations can be coordinated. We analyze several important statistical problems, including location estimation, linear regression, and non-parametric density estimation. 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. This holds for both independent and joint shifts, but the least favorable perturbations and minimax risks differ. For other problems, we provide nearly optimal estimators and precise finite-sample bounds. We also introduce several tools for bounding the minimax risk under distribution shift, such as a smoothing technique for location families, and generalizations of classical tools including least favorable sequences of priors, the modulus of continuity, Le Cam's, Fano's, and Assouad's methods.

* 60 pages, 7 figures 
Viaarxiv icon

Efficient and Multiply Robust Risk Estimation under General Forms of Dataset Shift

Jun 29, 2023
Hongxiang Qiu, Eric Tchetgen Tchetgen, Edgar Dobriban

Figure 1 for Efficient and Multiply Robust Risk Estimation under General Forms of Dataset Shift
Figure 2 for Efficient and Multiply Robust Risk Estimation under General Forms of Dataset Shift

Statistical machine learning methods often face the challenge of limited data available from the population of interest. One remedy is to leverage data from auxiliary source populations, which share some conditional distributions or are linked in other ways with the target domain. Techniques leveraging such \emph{dataset shift} conditions are known as \emph{domain adaptation} or \emph{transfer learning}. 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. In this paper, we study the general problem of efficiently estimating target population risk under various dataset shift conditions, leveraging semiparametric efficiency theory. We consider a general class of dataset shift conditions, which includes three popular conditions -- covariate, label and concept shift -- as special cases. We allow for partially non-overlapping support between the source and target populations. We develop efficient and multiply robust estimators along with a straightforward specification test of these dataset shift conditions. We also derive efficiency bounds for two other dataset shift conditions, posterior drift and location-scale shift. Simulation studies support the efficiency gains due to leveraging plausible dataset shift conditions.

* add acknowledgment to the source of the data 
Viaarxiv icon

Optimal Heterogeneous Collaborative Linear Regression and Contextual Bandits

Jun 09, 2023
Xinmeng Huang, Kan Xu, Donghwan Lee, Hamed Hassani, Hamsa Bastani, Edgar Dobriban

Figure 1 for Optimal Heterogeneous Collaborative Linear Regression and Contextual Bandits
Figure 2 for Optimal Heterogeneous Collaborative Linear Regression and Contextual Bandits
Figure 3 for Optimal Heterogeneous Collaborative Linear Regression and Contextual Bandits
Figure 4 for Optimal Heterogeneous Collaborative Linear Regression and Contextual Bandits

Large and complex datasets are often collected from several, possibly heterogeneous sources. Collaborative learning methods improve efficiency by leveraging commonalities across datasets while accounting for possible differences among them. Here we study collaborative linear regression and contextual bandits, where each instance's associated parameters are equal to a global parameter plus a sparse instance-specific term. We propose a novel two-stage estimator called MOLAR that leverages this structure by first constructing an entry-wise median of the instances' linear regression estimates, and then shrinking the instance-specific estimates towards the median. MOLAR improves the dependence of the estimation error on the data dimension, compared to independent least squares estimates. We then apply MOLAR to develop methods for sparsely heterogeneous collaborative contextual bandits, which lead to improved regret guarantees compared to independent bandit methods. We further show that our methods are minimax optimal by providing a number of lower bounds. Finally, we support the efficiency of our methods by performing experiments on both synthetic data and the PISA dataset on student educational outcomes from heterogeneous countries.

Viaarxiv icon

Sharp-SSL: Selective high-dimensional axis-aligned random projections for semi-supervised learning

Apr 18, 2023
Tengyao Wang, Edgar Dobriban, Milana Gataric, Richard J. Samworth

Figure 1 for Sharp-SSL: Selective high-dimensional axis-aligned random projections for semi-supervised learning
Figure 2 for Sharp-SSL: Selective high-dimensional axis-aligned random projections for semi-supervised learning
Figure 3 for Sharp-SSL: Selective high-dimensional axis-aligned random projections for semi-supervised learning
Figure 4 for 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. Our primary goal is to identify important variables for distinguishing between the classes; existing low-dimensional methods can then be applied for final class assignment. Motivated by a generalized Rayleigh quotient, we score projections according to the traces of the estimated whitened between-class covariance matrices on the projected data. This enables us to assign an importance weight to each variable for a given projection, and to select our signal variables by aggregating these weights over high-scoring projections. Our theory shows that the resulting Sharp-SSL algorithm is able to recover the signal coordinates with high probability when we aggregate over sufficiently many random projections and when the base procedure estimates the whitened between-class covariance matrix sufficiently well. The Gaussian EM algorithm is a natural choice as a base procedure, and we provide a new analysis of its performance in semi-supervised settings that controls the parameter estimation error in terms of the proportion of labeled data in the sample. Numerical results on both simulated data and a real colon tumor dataset support the excellent empirical performance of the method.

* 49 pages, 4 figures 
Viaarxiv icon

Demystifying Disagreement-on-the-Line in High Dimensions

Jan 31, 2023
Donghwan Lee, Behrad Moniri, Xinmeng Huang, Edgar Dobriban, Hamed Hassani

Figure 1 for Demystifying Disagreement-on-the-Line in High Dimensions
Figure 2 for Demystifying Disagreement-on-the-Line in High Dimensions
Figure 3 for Demystifying Disagreement-on-the-Line in High Dimensions
Figure 4 for 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. Recent work suggests that the notion of disagreement, the degree to which two models trained with different randomness differ on the same input, is a key to tackle this problem. Experimentally, disagreement and prediction error have been shown to be strongly connected, which has been used to estimate model performance. Experiments have lead to the discovery of the disagreement-on-the-line phenomenon, whereby the classification error under the target domain is often a linear function of the classification error under the source domain; and whenever this property holds, disagreement under the source and target domain follow the same linear relation. In this work, we develop a theoretical foundation for analyzing disagreement in high-dimensional random features regression; and study under what conditions the disagreement-on-the-line phenomenon occurs in our setting. Experiments on CIFAR-10-C, Tiny ImageNet-C, and Camelyon17 are consistent with our theory and support the universality of the theoretical findings.

Viaarxiv icon

Conformal Frequency Estimation with Sketched Data under Relaxed Exchangeability

Nov 09, 2022
Matteo Sesia, Stefano Favaro, Edgar Dobriban

Figure 1 for Conformal Frequency Estimation with Sketched Data under Relaxed Exchangeability
Figure 2 for Conformal Frequency Estimation with Sketched Data under Relaxed Exchangeability
Figure 3 for Conformal Frequency Estimation with Sketched Data under Relaxed Exchangeability
Figure 4 for Conformal Frequency Estimation with Sketched Data under Relaxed Exchangeability

A flexible method is developed to construct a confidence interval for the frequency of a queried object in a very large data set, based on a much smaller sketch of the data. The approach requires no knowledge of the data distribution or of the details of the sketching algorithm; instead, it constructs provably valid frequentist confidence intervals for random queries using a conformal inference approach. After achieving marginal coverage for random queries under the assumption of data exchangeability, the proposed method is extended to provide stronger inferences accounting for possibly heterogeneous frequencies of different random queries, redundant queries, and distribution shifts. While the presented methods are broadly applicable, this paper focuses on use cases involving the count-min sketch algorithm and a non-linear variation thereof, to facilitate comparison to prior work. In particular, the developed methods are compared empirically to frequentist and Bayesian alternatives, through simulations and experiments with data sets of SARS-CoV-2 DNA sequences and classic English literature.

* 56 pages, 31 figures, 2 tables. arXiv admin note: substantial text overlap with arXiv:2204.04270 
Viaarxiv icon

PAC Prediction Sets for Meta-Learning

Jul 06, 2022
Sangdon Park, Edgar Dobriban, Insup Lee, Osbert Bastani

Figure 1 for PAC Prediction Sets for Meta-Learning
Figure 2 for PAC Prediction Sets for Meta-Learning
Figure 3 for PAC Prediction Sets for Meta-Learning
Figure 4 for 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. We study this problem in the context of meta learning, where the goal is to quickly adapt a predictor to new tasks. In particular, we propose a novel algorithm to construct \emph{PAC prediction sets}, which capture uncertainty via sets of labels, that can be adapted to new tasks with only a few training examples. These prediction sets satisfy an extension of the typical PAC guarantee to the meta learning setting; in particular, the PAC guarantee holds with high probability over future tasks. We demonstrate the efficacy of our approach on four datasets across three application domains: mini-ImageNet and CIFAR10-C in the visual domain, FewRel in the language domain, and the CDC Heart Dataset in the medical domain. In particular, our prediction sets satisfy the PAC guarantee while having smaller size compared to other baselines that also satisfy this guarantee.

Viaarxiv icon