Search Results for author:
Found 80 papers, 23 papers with code
Improved Differentially Private Regression via Gradient Boosting
We revisit the problem of differentially private squared error linear regression.
Accuracy First: Selecting a Differential Privacy Level for Accuracy-Constrained ERM
Traditional approaches to differential privacy assume a fixed privacy requirement $\epsilon$ for a computation, and attempt to maximize the accuracy of the computation subject to the privacy constraint.
Gaussian Differential Privacy
More precisely, the privacy guarantees of \emph{any} hypothesis testing based definition of privacy (including original DP) converges to GDP in the limit under composition.
Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
We prove that the computational problem of auditing subgroup fairness for both equality of false positive rates and statistical parity is equivalent to the problem of weak agnostic learning, which means it is computationally hard in the worst case, even for simple structured subclasses.
An Empirical Study of Rich Subgroup Fairness for Machine Learning
In this paper, we undertake an extensive empirical evaluation of the algorithm of Kearns et al. On four real datasets for which fairness is a concern, we investigate the basic convergence of the algorithm when instantiated with fast heuristics in place of learning oracles, measure the tradeoffs between fairness and accuracy, and compare this approach with the recent algorithm of Agarwal et al. [2018], which implements weaker and more traditional marginal fairness constraints defined by individual protected attributes.
Adaptive Machine Unlearning
In this paper, we give a general reduction from deletion guarantees against adaptive sequences to deletion guarantees against non-adaptive sequences, using differential privacy and its connection to max information.
Differentially Private Query Release Through Adaptive Projection
We propose, implement, and evaluate a new algorithm for releasing answers to very large numbers of statistical queries like $k$-way marginals, subject to differential privacy.
Practical Adversarial Multivalid Conformal Prediction
It is computationally lightweight -- comparable to split conformal prediction -- but does not require having a held-out validation set, and so all data can be used for training models from which to derive a conformal score.
Batch Multivalid Conformal Prediction
Multivalid coverage guarantees are stronger than marginal coverage guarantees in two ways: (1) They hold even conditional on group membership -- that is, the target coverage level $1-\alpha$ holds conditionally on membership in each of an arbitrary (potentially intersecting) group in a finite collection $\mathcal{G}$ of regions in the feature space.
Confidence-Ranked Reconstruction of Census Microdata from Published Statistics
Our attacks significantly outperform those that are based only on access to a public distribution or population from which the private dataset $D$ was sampled, demonstrating that they are exploiting information in the aggregate statistics $Q(D)$, and not simply the overall structure of the distribution.
Descent-to-Delete: Gradient-Based Methods for Machine Unlearning
We study the data deletion problem for convex models.
Minimax Group Fairness: Algorithms and Experiments
We consider a recently introduced framework in which fairness is measured by worst-case outcomes across groups, rather than by the more standard differences between group outcomes.
Higher-Order Approximate Relational Refinement Types for Mechanism Design and Differential Privacy
Unlike typical programmatic properties, it is not sufficient for algorithms to merely satisfy the property---incentive properties are only useful if the strategic agents also believe this fact.
Programming Languages Computer Science and Game Theory
Computer-aided verification in mechanism design
To address both concerns, we explore techniques from computer-aided verification to construct formal proofs of incentive properties.
Computer Science and Game Theory Logic in Computer Science
Multicalibration as Boosting for Regression
Using this characterization, we give an exceedingly simple algorithm that can be analyzed both as a boosting algorithm for regression and as a multicalibration algorithm for a class H that makes use only of a standard squared error regression oracle for H. We give a weak learning assumption on H that ensures convergence to Bayes optimality without the need to make any realizability assumptions -- giving us an agnostic boosting algorithm for regression.
Oracle Efficient Private Non-Convex Optimization
We find that for the problem of learning linear classifiers, directly optimizing for 0/1 loss using our approach can out-perform the more standard approach of privately optimizing a convex-surrogate loss function on the Adult dataset.
Generalization in Adaptive Data Analysis and Holdout Reuse
We also formalize and address the general problem of data reuse in adaptive data analysis.
Algorithms and Learning for Fair Portfolio Design
We consider a variation on the classical finance problem of optimal portfolio design.
An Algorithmic Framework for Fairness Elicitation
We consider settings in which the right notion of fairness is not captured by simple mathematical definitions (such as equality of error rates across groups), but might be more complex and nuanced and thus require elicitation from individual or collective stakeholders.
Guaranteed Validity for Empirical Approaches to Adaptive Data Analysis
We design a general framework for answering adaptive statistical queries that focuses on providing explicit confidence intervals along with point estimates.
Mitigating Bias in Adaptive Data Gathering via Differential Privacy
Data that is gathered adaptively --- via bandit algorithms, for example --- exhibits bias.
Local Differential Privacy for Evolving Data
Moreover, existing techniques to mitigate this effect do not apply in the "local model" of differential privacy that these systems use.
Online Learning with an Unknown Fairness Metric
We consider the problem of online learning in the linear contextual bandits setting, but in which there are also strong individual fairness constraints governed by an unknown similarity metric.
A Smoothed Analysis of the Greedy Algorithm for the Linear Contextual Bandit Problem
Bandit learning is characterized by the tension between long-term exploration and short-term exploitation.
Strategic Classification from Revealed Preferences
We study an online linear classification problem, in which the data is generated by strategic agents who manipulate their features in an effort to change the classification outcome.
Fairness in Reinforcement Learning
We initiate the study of fairness in reinforcement learning, where the actions of a learning algorithm may affect its environment and future rewards.
Multidimensional Dynamic Pricing for Welfare Maximization
We are able to apply this technique to the setting of unit demand buyers despite the fact that in that setting the goods are not divisible, and the natural fractional relaxation of a unit demand valuation is not strongly concave.
A Convex Framework for Fair Regression
We introduce a flexible family of fairness regularizers for (linear and logistic) regression problems.
Fairness in Criminal Justice Risk Assessments: The State of the Art
Methods: We draw on the existing literatures in criminology, computer science and statistics to provide an integrated examination of fairness and accuracy in criminal justice risk assessments.
Fairness in Learning: Classic and Contextual Bandits
This tight connection allows us to provide a provably fair algorithm for the linear contextual bandit problem with a polynomial dependence on the dimension, and to show (for a different class of functions) a worst-case exponential gap in regret between fair and non-fair learning algorithms
Learning from Rational Behavior: Predicting Solutions to Unknown Linear Programs
This models the problem of predicting the behavior of a rational agent whose goals are known, but whose resources are unknown.
Max-Information, Differential Privacy, and Post-Selection Hypothesis Testing
In this paper, we initiate a principled study of how the generalization properties of approximate differential privacy can be used to perform adaptive hypothesis testing, while giving statistically valid $p$-value corrections.
Do Prices Coordinate Markets?
Second, we provide learning-theoretic results that show that such prices are robust to changing the buyers in the market, so long as all buyers are sampled from the same (unknown) distribution.
Adaptive Learning with Robust Generalization Guarantees
We also show that perfect generalization is a strictly stronger guarantee than differential privacy, but that, nevertheless, many learning tasks can be carried out subject to the guarantees of perfect generalization.
Preserving Statistical Validity in Adaptive Data Analysis
We show that, surprisingly, there is a way to estimate an exponential in $n$ number of expectations accurately even if the functions are chosen adaptively.
Dual Query: Practical Private Query Release for High Dimensional Data
We present a practical, differentially private algorithm for answering a large number of queries on high dimensional datasets.
Watch and Learn: Optimizing from Revealed Preferences Feedback
In this paper we present an approach to solving for the leader's optimal strategy in certain Stackelberg games where the follower's utility function (and thus the subsequent best response of the follower) is unknown.
Online Learning and Profit Maximization from Revealed Preferences
We consider the problem of learning from revealed preferences in an online setting.
Privately Solving Linear Programs
In this paper, we initiate the systematic study of solving linear programs under differential privacy.
Downstream Effects of Affirmative Action
We show that both goals can be achieved when the college does not report grades.
Fair Algorithms for Learning in Allocation Problems
We formalize this fairness notion for allocation problems and investigate its algorithmic consequences.
The Frontiers of Fairness in Machine Learning
The last few years have seen an explosion of academic and popular interest in algorithmic fairness.
How to Use Heuristics for Differential Privacy
We show that there is an efficient algorithm for privately constructing synthetic data for any such class, given a non-private learning oracle.
Differentially Private Fair Learning
This algorithm is appealingly simple, but must be able to use protected group membership explicitly at test time, which can be viewed as a form of 'disparate treatment'.
Accuracy First: Selecting a Differential Privacy Level for Accuracy Constrained ERM
Traditional approaches to differential privacy assume a fixed privacy requirement ε for a computation, and attempt to maximize the accuracy of the computation subject to the privacy constraint.
Meritocratic Fairness for Cross-Population Selection
We consider the problem of selecting a strong pool of individuals from several populations with incomparable skills (e. g. soccer players, mathematicians, and singers) in a fair manner.
Equal Opportunity in Online Classification with Partial Feedback
We study an online classification problem with partial feedback in which individuals arrive one at a time from a fixed but unknown distribution, and must be classified as positive or negative.
The Role of Interactivity in Local Differential Privacy
Next, we show that our reduction is tight by exhibiting a family of problems such that for any $k$, there is a fully interactive $k$-compositional protocol which solves the problem, while no sequentially interactive protocol can solve the problem without at least an $\tilde \Omega(k)$ factor more examples.
Average Individual Fairness: Algorithms, Generalization and Experiments
Given a sample of individuals and classification problems, we design an oracle-efficient algorithm (i. e. one that is given access to any standard, fairness-free learning heuristic) for the fair empirical risk minimization task.
Fuzzi: A Three-Level Logic for Differential Privacy
Curators of sensitive datasets sometimes need to know whether queries against the data are differentially private [Dwork et al. 2006].
Programming Languages Logic in Computer Science
Exponential Separations in Local Differential Privacy
We prove a general connection between the communication complexity of two-player games and the sample complexity of their multi-player locally private analogues.
A New Analysis of Differential Privacy's Generalization Guarantees
This second claim follows from a thought experiment in which we imagine that the dataset is resampled from the posterior distribution after the mechanism has committed to its answers.
Optimal, Truthful, and Private Securities Lending
We consider a fundamental dynamic allocation problem motivated by the problem of $\textit{securities lending}$ in financial markets, the mechanism underlying the short selling of stocks.
Pipeline Interventions
We consider two objectives: social welfare maximization, and a fairness-motivated maximin objective that seeks to maximize the value to the population (starting node) with the \emph{least} expected value.
Fair Prediction with Endogenous Behavior
There is increasing regulatory interest in whether machine learning algorithms deployed in consequential domains (e. g. in criminal justice) treat different demographic groups "fairly."
Differentially Private Combinatorial Optimization
Is it even possible to design good algorithms for this problem that preserve the privacy of the clients?
Data Structures and Algorithms Cryptography and Security Computer Science and Game Theory
Moment Multicalibration for Uncertainty Estimation
We show how to achieve the notion of "multicalibration" from H\'ebert-Johnson et al. [2018] not just for means, but also for variances and other higher moments.
Online Multivalid Learning: Means, Moments, and Prediction Intervals
We present a general, efficient technique for providing contextual predictions that are "multivalid" in various senses, against an online sequence of adversarially chosen examples $(x, y)$.
Lexicographically Fair Learning: Algorithms and Generalization
We extend the notion of minimax fairness in supervised learning problems to its natural conclusion: lexicographic minimax fairness (or lexifairness for short).
Rejoinder: Gaussian Differential Privacy
In this rejoinder, we aim to address two broad issues that cover most comments made in the discussion.
Multiaccurate Proxies for Downstream Fairness
The goal of the proxy is to allow a general "downstream" learner -- with minimal assumptions on their prediction task -- to be able to use the proxy to train a model that is fair with respect to the true sensitive features.
Online Minimax Multiobjective Optimization: Multicalibeating and Other Applications
We introduce a simple but general online learning framework in which a learner plays against an adversary in a vector-valued game that changes every round.
An Algorithmic Framework for Bias Bounties
We propose and analyze an algorithmic framework for "bias bounties": events in which external participants are invited to propose improvements to a trained model, akin to bug bounty events in software and security.
Mixed Differential Privacy in Computer Vision
AdaMix incorporates few-shot training, or cross-modal zero-shot learning, on public data prior to private fine-tuning, to improve the trade-off.
Individually Fair Learning with One-Sided Feedback
We consider an online learning problem with one-sided feedback, in which the learner is able to observe the true label only for positively predicted instances.
Reconciling Individual Probability Forecasts
Individual probabilities refer to the probabilities of outcomes that are realized only once: the probability that it will rain tomorrow, the probability that Alice will die within the next 12 months, the probability that Bob will be arrested for a violent crime in the next 18 months, etc.
Private Synthetic Data for Multitask Learning and Marginal Queries
A key innovation in our algorithm is the ability to directly handle numerical features, in contrast to a number of related prior approaches which require numerical features to be first converted into {high cardinality} categorical features via {a binning strategy}.
Multicalibrated Regression for Downstream Fairness
We show how to take a regression function $\hat{f}$ that is appropriately ``multicalibrated'' and efficiently post-process it into an approximately error minimizing classifier satisfying a large variety of fairness constraints.
The Scope of Multicalibration: Characterizing Multicalibration via Property Elicitation
To further counter-weigh our negative result, we show that if a property $\Gamma^1$ is not elicitable by itself, but is elicitable conditionally on another elicitable property $\Gamma^0$, then there is a canonical algorithm that jointly multicalibrates $\Gamma^1$ and $\Gamma^0$; this generalizes past work on mean-moment multicalibration.
Balanced Filtering via Non-Disclosive Proxies
Importantly, we require that the proxy classification itself not reveal significant information about the sensitive group membership of any individual sample (i. e., it should be sufficiently non-disclosive).
Oracle Efficient Online Multicalibration and Omniprediction
We develop a new online multicalibration algorithm that is well defined for infinite benchmark classes $F$, and is oracle efficient (i. e. for any class $F$, the algorithm has the form of an efficient reduction to a no-regret learning algorithm for $F$).
Oracle Efficient Algorithms for Groupwise Regret
Our approach gives similar regret guarantees compared to [Blum & Lykouris]; however, we run in time linear in the number of groups, and are oracle-efficient in the hypothesis class.
High-Dimensional Prediction for Sequential Decision Making
We study the problem of making predictions of an adversarially chosen high-dimensional state that are unbiased subject to an arbitrary collection of conditioning events, with the goal of tailoring these events to downstream decision makers.
Membership Inference Attacks on Diffusion Models via Quantile Regression
Our proposed MI attack learns quantile regression models that predict (a quantile of) the distribution of reconstruction loss on examples not used in training.
Forecasting for Swap Regret for All Downstream Agents
In the low dimensional setting, we show how to make predictions such that all agents who best respond to our predictions have diminishing swap regret -- in 1 dimension, at the optimal $O(\sqrt{T})$ rate.
Diversified Ensembling: An Experiment in Crowdsourced Machine Learning
There, unlike in classical crowdsourced ML, participants deliberately specialize their efforts by working on subproblems, such as demographic subgroups in the service of fairness.
An Elementary Predictor Obtaining $2\sqrt{T}$ Distance to Calibration
Blasiok et al. [2023] proposed distance to calibration as a natural measure of calibration error that unlike expected calibration error (ECE) is continuous.
Repeated Contracting with Multiple Non-Myopic Agents: Policy Regret and Limited Liability
First, we show that this game admits a pure-strategy \emph{non-responsive} equilibrium amongst the Agents -- informally an equilibrium in which the Agent's actions depend on the history of realized states of nature, but not on the history of each other's actions, and so avoids the complexities of collusion and threats.
Multicalibration for Confidence Scoring in LLMs
This paper proposes the use of "multicalibration" to yield interpretable and reliable confidence scores for outputs generated by large language models (LLMs).
