Agnostic Estimation of Mean and Covariance
Not on the map. Why?
Abstract (arXiv)
We consider the problem of estimating the mean and covariance of a distribution from iid samples in ℝn, in the presence of an η fraction of malicious noise; this is in contrast to much recent work where the noise itself is assumed to be from a distribution of known type. The agnostic problem includes many interesting special cases, e.g., learning the parameters of a single Gaussian (or finding the best-fit Gaussian) when η fraction of data is adversarially corrupted, agnostically learning a mixture of Gaussians, agnostic ICA, etc. We present polynomial-time algorithms to estimate the mean and covariance with error guarantees in terms of information-theoretic lower bounds. As a corollary, we also obtain an agnostic algorithm for Singular Value Decomposition.
Key facts. Cited by 131 papers in the map (data through 14 Oct 2026). First posted on arXiv 24 Apr 2016.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 35; 2022: 32; 2023: 21; 2024: 19; 2025: 15; 2026: 9.
131 papers.
Sort:
By area:
- 2021Remember What You Want to Forget: Algorithms for Machine Unlearningcited by 371
- 2021SPECTRE: Defending Against Backdoor Attacks Using Robust Statisticscited by 130
- 2021Robust and Differentially Private Mean Estimationcited by 66
- 2021Differential privacy and robust statistics in high dimensionscited by 64
- 2022Byzantine Machine Learning Made Easy by Resilient Averaging of Momentumscited by 63
- 2021Property Inference From Poisoningcited by 60
- 2021DORO: Distributional and Outlier Robust Optimizationcited by 56
- 2022Improved Certified Defenses against Data Poisoning with (Deterministic) Finite Aggregationcited by 52
- 2021Corruption-Robust Offline Reinforcement Learningcited by 51
- 2021Efficient Mean Estimation with Pure Differential Privacy via a Sum-of-Squares Exponential Mechanismcited by 50
- 2021Private and polynomial time algorithms for learning Gaussians and beyondcited by 39
- 2021Private Robust Estimation by Stabilizing Convex Relaxationscited by 38
38 more
- 2021Robust Policy Gradient against Strong Data Corruptioncited by 30
- 2021Byzantine-robust Federated Learning through Collaborative Malicious Gradient Filteringcited by 30
- 2022Byzantine-Robust Federated Learning with Optimal Statistical Rates and Privacy Guaranteescited by 30
- 2022On the Impossible Safety of Large AI Modelscited by 28
- 2021Statistical Query Lower Bounds for List-Decodable Linear Regressioncited by 25
- 2022Indiscriminate Data Poisoning Attacks on Neural Networkscited by 25
- 2022Robust Sparse Mean Estimation via Sum of Squarescited by 24
- 2021Clustering Mixture Models in Almost-Linear Time via List-Decodable Mean Estimationcited by 21
- 2021Efficient Algorithms for Learning from Coarse Labelscited by 21
- 2022Streaming Algorithms for High-Dimensional Robust Statisticscited by 21
- 2023Private estimation algorithms for stochastic block models and mixture modelscited by 21
- 2024The Utility and Complexity of in- and out-of-Distribution Machine Unlearningcited by 20
- 2023Run-Off Election: Improved Provable Defense against Data Poisoning Attackscited by 17
- 2021Robust Regression Revisited: Acceleration and Improved Estimation Ratescited by 15
- 2024Tolerant Algorithms for Learning with Arbitrary Covariate Shiftcited by 15
- 2021Outlier-Robust Learning of Ising Models Under Dobrushin's Conditioncited by 14
- 2021Outlier-Robust Sparse Estimation via Non-Convex Optimizationcited by 14
- 2022On Optimal Learning Under Targeted Data Poisoningcited by 14
- 2021Learning and Certification under Instance-targeted Poisoningcited by 13
- 2022Byzantine-Robust Online and Offline Distributed Reinforcement Learningcited by 13
- 2022List-Decodable Sparse Mean Estimation via Difference-of-Pairs Filteringcited by 13
- 2022Outlier-Robust Sparse Mean Estimation for Heavy-Tailed Distributionscited by 13
- 2021Learning GMMs with Nearly Optimal Robustness Guaranteescited by 12
- 2021On the power of adaptivity in statistical adversariescited by 12
- 2022Minimax Rates for Robust Community Detectioncited by 12
- 2021Sample-Optimal PAC Learning of Halfspaces with Malicious Noisecited by 11
- 2022List-Decodable Sparse Mean Estimationcited by 11
- 2022Estimation Contracts for Outlier-Robust Geometric Perceptioncited by 11
- 2021ReLU Regression with Massart Noisecited by 10
- 2022Secure Federated Clusteringcited by 10
- 2023Nearly-Linear Time and Streaming Algorithms for Outlier-Robust PCAcited by 10
- 2024SoS Certifiability of Subgaussian Distributions and its Algorithmic Applicationscited by 10
- 2024Adaptive and oblivious statistical adversaries are equivalentcited by 9
- 2021Robust Estimation for Random Graphscited by 8
- 2022List-Decodable Covariance Estimationcited by 8
- 2021SoS Degree Reduction with Applications to Clustering and Robust Moment Estimationcited by 7
- 2022Reducing Certified Regression to Certified Classification for General Poisoning Attackscited by 7
- 2023High Dimensional Distributed Gradient Descent with Arbitrary Number of Byzantine Attackerscited by 7
50 most cited of 131 All 131 (JSON)
- 2026Quantile-based Loss Filtering for Outlier-Robust Stochastic Gradient Descent11 Sep 2026
- 2026Precise sample covariance spectral norm error -- an RDT view16 Jul 2026
- 2026Robust Moment-Based Estimation via Spectral Gradient Reweighting26 May 2026
- 2026Algorithms with Polynomially-Improved Approximation Factors for the 2 rightarrow q Norm, and Applications24 May 2026
- 2026Robust Statistical Estimators with Bounded Empirical Sensitivity21 May 2026
- 2026High-Dimensional Gaussian Mean Estimation under Realizable Contamination17 Mar 2026
- 2026Sample Complexity Bounds for Robust Mean Estimation with Mean-Shift Contamination25 Feb 2026
- 2026Robust Processing and Learning: Principles, Methods, and Wireless Applications10 Feb 2026
- 2026Learning Mixture Models via Efficient High-dimensional Sparse Fourier Transforms8 Jan 2026
- 2025Learning Confidence Ellipsoids and Applications to Robust Subspace Recovery18 Dec 2025
- 2025Robust and Sparse Estimation of Unbounded Density Ratio under Heavy Contamination10 Dec 2025
- 2025High-Accuracy List-Decodable Mean Estimation21 Nov 2025
38 more
- 2025Information-Computation Tradeoffs for Noiseless Linear Regression with Oblivious Contamination12 Oct 2025
- 2025Linear Regression under Missing or Corrupted Coordinates23 Sep 2025
- 2025Corruption-Tolerant Asynchronous Q-Learning with Near-Optimal Rates10 Sep 2025
- 2025On the Learnability of Distribution Classes with Adaptive Adversaries5 Sep 2025
- 2025Robust Estimation Under Heterogeneous Corruption Rates20 Aug 2025
- 2025Agnostic Learning under Targeted Poisoning: Optimal Rates and the Role of Randomness3 Jun 2025
- 2025Robust Federated Personalised Mean Estimation for the Gaussian Mixture Model28 Apr 2025
- 2025Learning High-dimensional Gaussians from Censored Data28 Apr 2025
- 2025Computing High-dimensional Confidence Sets for Arbitrary Distributions3 Apr 2025
- 2025Batch List-Decodable Linear Regression via Higher Moments12 Mar 2025
- 2025Efficient Multivariate Robust Mean Estimation Under Mean-Shift Contamination20 Feb 2025
- 2025Entangled Mean Estimation in High-Dimensions9 Jan 2025
- 2024SoS Certificates for Sparse Singular Values and Their Applications: Robust Statistics, Subspace Distortion, and More30 Dec 2024
- 2024The Utility and Complexity of in- and out-of-Distribution Machine Unlearning12 Dec 2024
- 2024The Broader Landscape of Robustness in Algorithmic Statistics3 Dec 2024
- 2024SoS Certifiability of Subgaussian Distributions and its Algorithmic Applications28 Oct 2024
- 2024Robust Estimation for Kernel Exponential Families with Smoothed Total Variation Distances28 Oct 2024
- 2024Optimal Robust Estimation under Local and Global Corruptions: Stronger Adversary and Smaller Error22 Oct 2024
- 2024Adaptive and oblivious statistical adversaries are equivalent17 Oct 2024
- 2024Robust Gradient Descent for Phase Retrieval14 Oct 2024
- 2024Data Distribution Valuation6 Oct 2024
- 2024Efficient Statistics With Unknown Truncation, Polynomial Time Algorithms, Beyond Gaussians2 Oct 2024
- 2024Efficient PAC Learning of Halfspaces with Constant Malicious Noise Rate2 Oct 2024
- 2024Distribution Learnability and Robustness25 Jun 2024
- 2024Tolerant Algorithms for Learning with Arbitrary Covariate Shift4 Jun 2024
- 2024Distribution Learning Meets Graph Structure Sampling13 May 2024
- 2024VALID: a Validated Algorithm for Learning in Decentralized Networks with Possible Adversarial Presence12 May 2024
- 2024Multigroup Robustness1 May 2024
- 2024MAC Advice for Facility Location Mechanism Design18 Mar 2024
- 2024Robust Second-Order Nonconvex Optimization and Its Application to Low Rank Matrix Sensing12 Mar 2024
- 2024A Sub-Quadratic Time Algorithm for Robust Sparse Mean Estimation7 Mar 2024
- 2023Balancing Privacy, Robustness, and Efficiency in Machine Learning22 Dec 2023
- 2023Attacking Byzantine Robust Aggregation in High Dimensions22 Dec 2023
- 2023Clustering Mixtures of Bounded Covariance Distributions Under Optimal Separation19 Dec 2023
- 2023Near-Optimal Mean Estimation with Unknown, Heteroskedastic Variances5 Dec 2023
- 2023Near-Optimal Algorithms for Gaussians with Huber Contamination: Mean Estimation and Linear Regression4 Dec 2023
- 2023Online Robust Mean Estimation24 Oct 2023
- 2023Byzantine-Resilient Federated PCA and Low Rank Column-wise Sensing25 Sep 2023
50 newest of 131 All 131 (JSON)
- 2021Remember What You Want to Forget: Algorithms for Machine Unlearningcited by 371
- 2022Improved Certified Defenses against Data Poisoning with (Deterministic) Finite Aggregationcited by 52
- 2022Indiscriminate Data Poisoning Attacks on Neural Networkscited by 25
- 2024The Utility and Complexity of in- and out-of-Distribution Machine Unlearningcited by 20
- 2023Run-Off Election: Improved Provable Defense against Data Poisoning Attackscited by 17
- 2022On Optimal Learning Under Targeted Data Poisoningcited by 14
- 2024The Utility and Complexity of in- and out-of-Distribution Machine Unlearning12 Dec 2024
- 2023Run-Off Election: Improved Provable Defense against Data Poisoning Attacks5 Feb 2023
- 2022On Optimal Learning Under Targeted Data Poisoning6 Oct 2022
- 2022Indiscriminate Data Poisoning Attacks on Neural Networks19 Apr 2022
- 2022Improved Certified Defenses against Data Poisoning with (Deterministic) Finite Aggregation5 Feb 2022
- 2021Remember What You Want to Forget: Algorithms for Machine Unlearning4 Mar 2021
- 2021Corruption-Robust Offline Reinforcement Learningcited by 51
- 2021Corruption-Robust Offline Reinforcement Learning11 Jun 2021
- 2021SPECTRE: Defending Against Backdoor Attacks Using Robust Statisticscited by 130
- 2021Robust and Differentially Private Mean Estimationcited by 66
- 2021Differential privacy and robust statistics in high dimensionscited by 64
- 2022Byzantine Machine Learning Made Easy by Resilient Averaging of Momentumscited by 63
- 2021Property Inference From Poisoningcited by 60
- 2021DORO: Distributional and Outlier Robust Optimizationcited by 56
- 2021Efficient Mean Estimation with Pure Differential Privacy via a Sum-of-Squares Exponential Mechanismcited by 50
- 2021Private and polynomial time algorithms for learning Gaussians and beyondcited by 39
- 2021Private Robust Estimation by Stabilizing Convex Relaxationscited by 38
- 2021Robust Policy Gradient against Strong Data Corruptioncited by 30
- 2021Byzantine-robust Federated Learning through Collaborative Malicious Gradient Filteringcited by 30
- 2022Byzantine-Robust Federated Learning with Optimal Statistical Rates and Privacy Guaranteescited by 30
12 most cited of 123 All 123 (JSON)
- 2026Quantile-based Loss Filtering for Outlier-Robust Stochastic Gradient Descent11 Sep 2026
- 2026Precise sample covariance spectral norm error -- an RDT view16 Jul 2026
- 2026Robust Moment-Based Estimation via Spectral Gradient Reweighting26 May 2026
- 2026Algorithms with Polynomially-Improved Approximation Factors for the 2 rightarrow q Norm, and Applications24 May 2026
- 2026Robust Statistical Estimators with Bounded Empirical Sensitivity21 May 2026
- 2026High-Dimensional Gaussian Mean Estimation under Realizable Contamination17 Mar 2026
- 2026Sample Complexity Bounds for Robust Mean Estimation with Mean-Shift Contamination25 Feb 2026
- 2026Robust Processing and Learning: Principles, Methods, and Wireless Applications10 Feb 2026
- 2026Learning Mixture Models via Efficient High-dimensional Sparse Fourier Transforms8 Jan 2026
- 2025Learning Confidence Ellipsoids and Applications to Robust Subspace Recovery18 Dec 2025
- 2025Robust and Sparse Estimation of Unbounded Density Ratio under Heavy Contamination10 Dec 2025
- 2025High-Accuracy List-Decodable Mean Estimation21 Nov 2025
12 newest of 123 All 123 (JSON)