Search Results for author:
Found 109 papers, 29 papers with code
Differentially Private Reinforcement Learning with Self-Play
We study the problem of multi-agent reinforcement learning (multi-agent RL) with differential privacy (DP) constraints.
CPR: Retrieval Augmented Generation for Copyright Protection
To reduce risks of leaking private information contained in the retrieved set, we introduce Copy-Protected generation with Retrieval (CPR), a new method for RAG with strong copyright protection guarantees in a mixed-private setting for diffusion models. CPR allows to condition the output of diffusion models on a set of retrieved images, while also guaranteeing that unique identifiable information about those example is not exposed in the generated outputs.
Privacy Profiles for Private Selection
Private selection mechanisms (e. g., Report Noisy Max, Sparse Vector) are fundamental primitives of differentially private (DP) data analysis with wide applications to private query release, voting, and hyperparameter tuning.
Permute-and-Flip: An optimally robust and watermarkable decoder for LLMs
In this paper, we propose a new decoding method called Permute-and-Flip (PF) decoder.
Online Feature Updates Improve Online (Generalized) Label Shift Adaptation
This paper addresses the prevalent issue of label shift in an online setting with missing labels, where data distributions change over time and obtaining timely labels is challenging.
Near-Optimal Reinforcement Learning with Self-Play under Adaptivity Constraints
We study the problem of multi-agent reinforcement learning (MARL) with adaptivity constraints -- a new problem motivated by real-world applications where deployments of new policies are costly and the number of policy updates must be minimized.
Weak-to-Strong Jailbreaking on Large Language Models
In this paper, we propose the weak-to-strong jailbreaking attack, an efficient method to attack aligned LLMs to produce harmful text.
Improving the Privacy and Practicality of Objective Perturbation for Differentially Private Linear Learners
In the arena of privacy-preserving machine learning, differentially private stochastic gradient descent (DP-SGD) has outstripped the objective perturbation mechanism in popularity and interest.
Pricing with Contextual Elasticity and Heteroscedastic Valuation
We study an online contextual dynamic pricing problem, where customers decide whether to purchase a product based on its features and price.
Communication-Efficient Federated Non-Linear Bandit Optimization
Federated optimization studies the problem of collaborative function optimization among multiple clients (e. g. mobile devices or organizations) under the coordination of a central server.
On the accuracy and efficiency of group-wise clipping in differentially private optimization
Recent advances have substantially improved the accuracy, memory cost, and training speed of differentially private (DP) deep learning, especially on large vision and language models with millions to billions of parameters.
Tractable MCMC for Private Learning with Pure and Gaussian Differential Privacy
Posterior sampling, i. e., exponential mechanism to sample from the posterior distribution, provides $\varepsilon$-pure differential privacy (DP) guarantees and does not suffer from potentially unbounded privacy breach introduced by $(\varepsilon,\delta)$-approximate DP.
Coupling public and private gradient provably helps optimization
The success of large neural networks is crucially determined by the availability of data.
Threshold KNN-Shapley: A Linear-Time and Privacy-Friendly Approach to Data Valuation
Data valuation aims to quantify the usefulness of individual data sources in training machine learning (ML) models, and is a critical aspect of data-centric ML research.
Model-Free Algorithm with Improved Sample Efficiency for Zero-Sum Markov Games
In this work, we propose a model-free stage-based Q-learning algorithm and show that it achieves the same sample complexity as the best model-based algorithm, and hence for the first time demonstrate that model-free algorithms can enjoy the same optimality in the $H$ dependence as model-based algorithms.
Nonparametric Classification on Low Dimensional Manifolds using Overparameterized Convolutional Residual Networks
Convolutional residual neural networks (ConvResNets), though overparameterized, can achieve remarkable prediction performance in practice, which cannot be well explained by conventional wisdom.
Provable Robust Watermarking for AI-Generated Text
We propose a robust and high-quality watermark method, Unigram-Watermark, by extending an existing approach with a simplified fixed grouping strategy.
Offline Policy Evaluation for Reinforcement Learning with Adaptively Collected Data
Developing theoretical guarantees on the sample complexity of offline RL methods is an important step towards making data-hungry RL algorithms practically viable.
"Private Prediction Strikes Back!'' Private Kernelized Nearest Neighbors with Individual Renyi Filter
Most existing approaches of differentially private (DP) machine learning focus on private training.
Invisible Image Watermarks Are Provably Removable Using Generative AI
However, if we do not require the watermarked image to look the same as the original one, watermarks that keep the image semantically similar can be an alternative defense against our attack.
Non-stationary Reinforcement Learning under General Function Approximation
To the best of our knowledge, this is the first dynamic regret analysis in non-stationary MDPs with general function approximation.
Improved Differentially Private Regression via Gradient Boosting
We revisit the problem of differentially private squared error linear regression.
No-Regret Linear Bandits beyond Realizability
It achieves a near-optimal $\sqrt{T}$ regret for problems that the best-known regret is almost linear in time horizon $T$.
Logarithmic Switching Cost in Reinforcement Learning beyond Linear MDPs
In many real-life reinforcement learning (RL) problems, deploying new policies is costly.
Protecting Language Generation Models via Invisible Watermarking
We can then detect the secret message by probing a suspect model to tell if it is distilled from the protected one.
Generalized PTR: User-Friendly Recipes for Data-Adaptive Algorithms with Differential Privacy
The ''Propose-Test-Release'' (PTR) framework is a classic recipe for designing differentially private (DP) algorithms that are data-adaptive, i. e. those that add less noise when the input dataset is nice.
Near-Optimal Differentially Private Reinforcement Learning
We close this gap for the JDP case by designing an $\epsilon$-JDP algorithm with a regret of $\widetilde{O}(\sqrt{SAH^2T}+S^2AH^3/\epsilon)$ which matches the information-theoretic lower bound of non-private learning for all choices of $\epsilon> S^{1. 5}A^{0. 5} H^2/\sqrt{T}$.
Offline Reinforcement Learning with Closed-Form Policy Improvement Operators
Behavior constrained policy optimization has been demonstrated to be a successful paradigm for tackling Offline Reinforcement Learning.
Global Optimization with Parametric Function Approximation
We consider the problem of global optimization with noisy zeroth order oracles - a well-motivated problem useful for various applications ranging from hyper-parameter tuning for deep learning to new material design.
Distillation-Resistant Watermarking for Model Protection in NLP
We prove that a protected model still retains the original accuracy within a certain bound.
Offline Reinforcement Learning with Differentiable Function Approximation is Provably Efficient
Offline reinforcement learning, which aims at optimizing sequential decision-making strategies with historical data, has been extensively applied in real-life applications.
Near-Optimal Deployment Efficiency in Reward-Free Reinforcement Learning with Linear Function Approximation
We study the problem of deployment efficient reinforcement learning (RL) with linear function approximation under the \emph{reward-free} exploration setting.
Differentially Private Optimization on Large Model at Small Cost
Our implementation achieves state-of-the-art (SOTA) accuracy with very small extra cost: on GPT2 and at almost the same memory cost (<1% overhead), BK has 1. 03X the time complexity of the standard training (0. 83X training speed in practice), and 0. 61X the time complexity of the most efficient DP implementation (1. 36X training speed in practice).
Differentially Private Bias-Term only Fine-tuning of Foundation Models
We study the problem of differentially private (DP) fine-tuning of large pre-trained models -- a recent privacy-preserving approach suitable for solving downstream tasks with sensitive data.
Doubly Fair Dynamic Pricing
We show that a doubly fair policy must be random to have higher revenue than the best trivial policy that assigns the same price to different groups.
Optimal Dynamic Regret in LQR Control
We consider the problem of nonstochastic control with a sequence of quadratic losses, i. e., LQR control.
Why Quantization Improves Generalization: NTK of Binary Weight Neural Networks
We propose a quasi neural network to approximate the distribution propagation, which is a neural network with continuous parameters and smooth activation function.
Offline Stochastic Shortest Path: Learning, Evaluation and Towards Optimality
Goal-oriented Reinforcement Learning, where the agent needs to reach the goal state while simultaneously minimizing the cost, has received significant attention in real-world applications.
Provably Confidential Language Modelling
Large language models are shown to memorize privacy information such as social security numbers in training data.
Second Order Path Variationals in Non-Stationary Online Learning
We consider the problem of universal dynamic regret minimization under exp-concave and smooth losses.
Deep Learning meets Nonparametric Regression: Are Weight-Decayed DNNs Locally Adaptive?
We consider a "Parallel NN" variant of deep ReLU networks and show that the standard weight decay is equivalent to promoting the $\ell_p$-sparsity ($0<p<1$) of the coefficient vector of an end-to-end learned function bases, i. e., a dictionary.
Towards Differential Relational Privacy and its use in Question Answering
While bounding general memorization can have detrimental effects on the performance of a trained model, bounding RM does not prevent effective learning.
Adaptive Private-K-Selection with Adaptive K and Application to Multi-label PATE
We provide an end-to-end Renyi DP based-framework for differentially private top-$k$ selection.
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.
Near-optimal Offline Reinforcement Learning with Linear Representation: Leveraging Variance Information with Pessimism
However, a precise understanding of the statistical limits with function representations, remains elusive, even when such a representation is linear.
Sample-Efficient Reinforcement Learning with loglog(T) Switching Cost
In this paper, we propose a new algorithm based on stage-wise exploration and adaptive policy elimination that achieves a regret of $\widetilde{O}(\sqrt{H^4S^2AT})$ while requiring a switching cost of $O(HSA \log\log T)$.
Towards Agnostic Feature-based Dynamic Pricing: Linear Policies vs Linear Valuation with Unknown Noise
In feature-based dynamic pricing, a seller sets appropriate prices for a sequence of products (described by feature vectors) on the fly by learning from the binary outcomes of previous sales sessions ("Sold" if valuation $\geq$ price, and "Not Sold" otherwise).
Optimal Dynamic Regret in Proper Online Learning with Strongly Convex Losses and Beyond
We study the framework of universal dynamic regret minimization with strongly convex losses.
Multivariate Trend Filtering for Lattice Data
We study a multivariate version of trend filtering, called Kronecker trend filtering or KTF, for the case in which the design points form a lattice in $d$ dimensions.
Privately Publishable Per-instance Privacy
We consider how to privately share the personalized privacy losses incurred by objective perturbation, using per-instance differential privacy (pDP).
Towards Instance-Optimal Offline Reinforcement Learning with Pessimism
We study the offline reinforcement learning (offline RL) problem, where the goal is to learn a reward-maximizing policy in an unknown Markov Decision Process (MDP) using the data coming from a policy $\mu$.
SeqPATE: Differentially Private Text Generation via Knowledge Distillation
Differentially private (DP) learning algorithms provide guarantees on identifying the existence of a training sample from model outputs.
Smoothed Differential Privacy
Differential privacy (DP) is a widely-accepted and widely-applied notion of privacy based on worst-case analysis.
Optimal Accounting of Differential Privacy via Characteristic Function
Characterizing the privacy degradation over compositions, i. e., privacy accounting, is a fundamental topic in differential privacy (DP) with many applications to differentially private machine learning and federated learning.
Optimal Uniform OPE and Model-based Offline Reinforcement Learning in Time-Homogeneous, Reward-Free and Task-Agnostic Settings
This work studies the statistical limits of uniform convergence for offline policy evaluation (OPE) problems with model-based methods (for episodic MDP) and provides a unified framework towards optimal learning for several well-motivated offline tasks.
Optimal Dynamic Regret in Exp-Concave Online Learning
We consider the problem of the Zinkevich (2003)-style dynamic regret minimization in online learning with exp-concave losses.
Logarithmic Regret in Feature-based Dynamic Pricing
Feature-based dynamic pricing is an increasingly popular model of setting prices for highly differentiated products with applications in digital marketing, online sales, real estate and so on.
Non-stationary Online Learning with Memory and Non-stochastic Control
We study the problem of Online Convex Optimization (OCO) with memory, which allows loss functions to depend on past decisions and thus captures temporal effects of learning problems.
Near-Optimal Offline Reinforcement Learning via Double Variance Reduction
Our main result shows that OPDVR provably identifies an $\epsilon$-optimal policy with $\widetilde{O}(H^2/d_m\epsilon^2)$ episodes of offline data in the finite-horizon stationary transition setting, where $H$ is the horizon length and $d_m$ is the minimal marginal state-action distribution induced by the behavior policy.
An Optimal Reduction of TV-Denoising to Adaptive Online Learning
We consider the problem of estimating a function from $n$ noisy samples whose discrete Total Variation (TV) is bounded by $C_n$.
Improving Sparse Vector Technique with Renyi Differential Privacy
The Sparse Vector Technique (SVT) is one of the most fundamental algorithmic tools in differential privacy (DP).
Revisiting Model-Agnostic Private Learning: Faster Rates and Active Learning
The Private Aggregation of Teacher Ensembles (PATE) framework is one of the most promising recent approaches in differentially private learning.
Inter-Series Attention Model for COVID-19 Forecasting
COVID-19 pandemic has an unprecedented impact all over the world since early 2020.
Voting-based Approaches For Differentially Private Federated Learning
Differentially Private Federated Learning (DPFL) is an emerging field with many applications.
Adaptive Online Estimation of Piecewise Polynomial Trends
We consider the framework of non-stationary stochastic optimization [Besbes et al, 2015] with squared error losses and noisy gradient feedback where the dynamic regret of an online learner against a time varying comparator sequence is studied.
Near-Optimal Provable Uniform Convergence in Offline Policy Evaluation for Reinforcement Learning
The problem of Offline Policy Evaluation (OPE) in Reinforcement Learning (RL) is a critical step towards applying RL in real-life applications.
Bullseye Polytope: A Scalable Clean-Label Poisoning Attack with Improved Transferability
Our attack, Bullseye Polytope, improves the attack success rate of the current state-of-the-art by 26. 75% in end-to-end transfer learning, while increasing attack speed by a factor of 12.
Domain Adaptation with Conditional Distribution Matching and Generalized Label Shift
However, recent work has shown limitations of this approach when label distributions differ between the source and target domains.
Asymptotically Efficient Off-Policy Evaluation for Tabular Reinforcement Learning
We consider the problem of off-policy evaluation for reinforcement learning, where the goal is to estimate the expected reward of a target policy $\pi$ using offline data collected by running a logging policy $\mu$.
Semantic Guided and Response Times Bounded Top-k Similarity Search over Knowledge Graphs
We face two challenges on graph query over a knowledge graph: (1) the structural gap between $G_Q$ and the predefined schema in $G$ causes mismatch with query graph, (2) users cannot view the answers until the graph query terminates, leading to a longer system response time (SRT).
Databases
Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting
Time series forecasting is an important problem across many domains, including predictions of solar plant energy output, electricity consumption, and traffic jam situation.
Ranked #27 on
Image Generation
on ImageNet 64x64
(Bits per dim metric)
Towards Optimal Off-Policy Evaluation for Reinforcement Learning with Marginalized Importance Sampling
To solve this problem, we consider a marginalized importance sampling (MIS) estimator that recursively estimates the state marginal distribution for the target policy at every step.
Doubly Robust Crowdsourcing
Large-scale labeled dataset is the indispensable fuel that ignites the AI revolution as we see today.
Online Forecasting of Total-Variation-bounded Sequences
We design an $O(n\log n)$-time algorithm that achieves a cumulative square error of $\tilde{O}(n^{1/3}C_n^{2/3}\sigma^{4/3} + C_n^2)$ with high probability. We also prove a lower bound that matches the upper bound in all parameters (up to a $\log(n)$ factor).
Provably Efficient Q-Learning with Low Switching Cost
We take initial steps in studying PAC-MDP algorithms with limited adaptivity, that is, algorithms that change its exploration policy as infrequently as possible during regret minimization.
A Higher-Order Kolmogorov-Smirnov Test
We present an extension of the Kolmogorov-Smirnov (KS) two-sample test, which can be more sensitive to differences in the tails.
Imitation-Regularized Offline Learning
To solve both problems, we show how one can use policy improvement (PIL) objectives, regularized by policy imitation (IML).
ProxQuant: Quantized Neural Networks via Proximal Operators
To make deep neural networks feasible in resource-constrained environments (such as mobile devices), it is beneficial to quantize models by using low-precision weights.
Subsampled Rényi Differential Privacy and Analytical Moments Accountant
We study the problem of subsampling in differential privacy (DP), a question that is the centerpiece behind many successful differentially private machine learning algorithms.
Patch-Based Image Hallucination for Super Resolution with Detail Reconstruction from Similar Sample Images
Image hallucination and super-resolution have been studied for decades, and many approaches have been proposed to upsample low-resolution images using information from the images themselves, multiple example images, or large image databases.
An end-to-end Differentially Private Latent Dirichlet Allocation Using a Spectral Algorithm
Overall, by combining the sensitivity and utility characterization, we obtain an end-to-end differentially private spectral algorithm for LDA and identify the corresponding configuration that outperforms others in any specific regime.
Improving the Gaussian Mechanism for Differential Privacy: Analytical Calibration and Optimal Denoising
The Gaussian mechanism is an essential building block used in multitude of differentially private data analysis algorithms.
Revisiting differentially private linear regression: optimal and adaptive prediction & estimation in unbounded domain
We revisit the problem of linear regression under a differential privacy constraint.
signSGD: Compressed Optimisation for Non-Convex Problems
Using a theorem by Gauss we prove that majority vote can achieve the same reduction in variance as full precision distributed SGD.
Detecting and Correcting for Label Shift with Black Box Predictors
Faced with distribution shift between training and test set, we wish to detect and quantify the shift, and to correct our classifiers without test set labels.
Convergence rate of sign stochastic gradient descent for non-convex functions
The sign stochastic gradient descent method (signSGD) utilizes only the sign of the stochastic gradient in its updates.
Higher-Order Total Variation Classes on Grids: Minimax Theory and Trend Filtering Methods
To move past this, we define two new higher-order TV classes, based on two ways of compiling the discrete derivatives of a parameter across the nodes.
Fully Convolutional Measurement Network for Compressive Sensing Image Reconstruction
Recently, deep learning methods have made a significant improvement in compressive sensing image reconstruction task.
Adaptive Measurement Network for CS Image Reconstruction
In this paper, we propose an adaptive measurement network in which measurement is obtained by learning.
Non-stationary Stochastic Optimization under $L_{p,q}$-Variation Measures
We consider a non-stationary sequential stochastic optimization problem, in which the underlying cost functions change over time under a variation budget constraint.
Per-instance Differential Privacy
We consider a refinement of differential privacy --- per instance differential privacy (pDP), which captures the privacy of a specific individual with respect to a fixed data set.
Optimal and Adaptive Off-policy Evaluation in Contextual Bandits
We study the off-policy evaluation problem---estimating the value of a target policy using data collected by another policy---under the contextual bandit model.
Understanding the 2016 US Presidential Election using ecological inference and distribution regression with census microdata
We combine fine-grained spatially referenced census data with the vote outcomes from the 2016 US presidential election.
Attributing Hacks
That is, given properties of sites and the temporal occurrence of attacks, we are able to attribute individual attacks to joint causes and vulnerabilities, as well as estimating the evolution of these vulnerabilities over time.
A Theoretical Analysis of Noisy Sparse Subspace Clustering on Dimensionality-Reduced Data
Subspace clustering is the problem of partitioning unlabeled data points into a number of clusters so that data points within one cluster lie approximately on a low-dimensional linear subspace.
Total Variation Classes Beyond 1d: Minimax Rates, and the Limitations of Linear Smoothers
Lastly, we investigate the problem of adaptivity of the total variation denoiser to these smaller Sobolev function spaces.
On-Average KL-Privacy and its equivalence to Generalization for Max-Entropy Mechanisms
We define On-Average KL-Privacy and present its properties and connections to differential privacy, generalization and information-theoretic quantities including max-information and mutual information.
A Minimax Theory for Adaptive Data Analysis
In this paper, we propose a minimax framework for adaptive data analysis.
Differentially private subspace clustering
Subspace clustering is an unsupervised learning problem that aims at grouping data points into multiple ``clusters'' so that data points in a single cluster lie approximately on a low-dimensional linear subspace.
Fast Differentially Private Matrix Factorization
Differentially private collaborative filtering is a challenging task, both in terms of accuracy and speed.
Graph Connectivity in Noisy Sparse Subspace Clustering
A line of recent work (4, 19, 24, 20) provided strong theoretical guarantee for sparse subspace clustering (4), the state-of-the-art algorithm for subspace clustering, on both noiseless and noisy data sets.
Privacy for Free: Posterior Sampling and Stochastic Gradient Monte Carlo
We consider the problem of Bayesian learning on sensitive datasets and present two simple but somewhat surprising results that connect Bayesian learning to "differential privacy:, a cryptographic approach to protect individual-level privacy while permiting database-level utility.
Learning with Differential Privacy: Stability, Learnability and the Sufficiency and Necessity of ERM Principle
Lastly, we extend some of the results to the more practical $(\epsilon,\delta)$-differential privacy and establish the existence of a phase-transition on the class of problems that are approximately privately learnable with respect to how small $\delta$ needs to be.
Trend Filtering on Graphs
We introduce a family of adaptive estimators on graphs, based on penalizing the $\ell_1$ norm of discrete graph differences.
Parallel and Distributed Block-Coordinate Frank-Wolfe Algorithms
We develop parallel and distributed Frank-Wolfe algorithms; the former on shared memory machines with mini-batching, and the latter in a delayed update framework.
The Falling Factorial Basis and Its Statistical Applications
We study a novel spline-like basis, which we name the "falling factorial basis", bearing many similarities to the classic truncated power basis.
Provable Subspace Clustering: When LRR meets SSC
Sparse Subspace Clustering (SSC) and Low-Rank Representation (LRR) are both considered as the state-of-the-art methods for {\em subspace clustering}.
Practical Matrix Completion and Corruption Recovery using Proximal Alternating Robust Subspace Minimization
Low-rank matrix completion is a problem of immense practical importance.
Noisy Sparse Subspace Clustering
This paper considers the problem of subspace clustering under noise.
