Higher degree sum-of-squares relaxations robust against oblivious outliers

• URL: http://arxiv.org/abs/2211.07327v1
• Date: Mon, 14 Nov 2022 13:09:12 GMT
• Title: Higher degree sum-of-squares relaxations robust against oblivious outliers
• Authors: Tommaso d'Orsi, Rajai Nasser, Gleb Novikov, David Steurer
• Abstract summary: We consider estimation models of the form $Y=X*+N$, where $X*$ is some $m$-dimensional signal we wish to recover. We introduce a family of algorithms that under mild assumptions recover the signal $X*$ in all estimation problems for which there exists a sum-of-squares algorithm.
• Score: 14.58610686291355
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We consider estimation models of the form $Y=X^*+N$, where $X^*$ is some $m$-dimensional signal we wish to recover, and $N$ is symmetrically distributed noise that may be unbounded in all but a small $\alpha$ fraction of the entries. We introduce a family of algorithms that under mild assumptions recover the signal $X^*$ in all estimation problems for which there exists a sum-of-squares algorithm that succeeds in recovering the signal $X^*$ when the noise $N$ is Gaussian. This essentially shows that it is enough to design a sum-of-squares algorithm for an estimation problem with Gaussian noise in order to get the algorithm that works with the symmetric noise model. Our framework extends far beyond previous results on symmetric noise models and is even robust to adversarial perturbations. As concrete examples, we investigate two problems for which no efficient algorithms were known to work for heavy-tailed noise: tensor PCA and sparse PCA. For the former, our algorithm recovers the principal component in polynomial time when the signal-to-noise ratio is at least $\tilde{O}(n^{p/4}/\alpha)$, that matches (up to logarithmic factors) current best known algorithmic guarantees for Gaussian noise. For the latter, our algorithm runs in quasipolynomial time and matches the state-of-the-art guarantees for quasipolynomial time algorithms in the case of Gaussian noise. Using a reduction from the planted clique problem, we provide evidence that the quasipolynomial time is likely to be necessary for sparse PCA with symmetric noise. In our proofs we use bounds on the covering numbers of sets of pseudo-expectations, which we obtain by certifying in sum-of-squares upper bounds on the Gaussian complexities of sets of solutions. This approach for bounding the covering numbers of sets of pseudo-expectations may be interesting in its own right and may find other application in future works.

Related papers

• Robust Sparse Regression with Non-Isotropic Designs [4.964650614497048]
  We develop a technique to design efficiently computable estimators for sparse linear regression in the simultaneous presence of two adversaries. We provide a novel analysis of weighted penalized Huber loss that is suitable for heavy-tailed designs in the presence of two adversaries.
  arXiv  Detail & Related papers  (2024-10-31T13:51:59Z)
• Perturb-and-Project: Differentially Private Similarities and Marginals [73.98880839337873]
  We revisit the input perturbations framework for differential privacy where noise is added to the input $Ain mathcalS$. We first design novel efficient algorithms to privately release pair-wise cosine similarities. We derive a novel algorithm to compute $k$-way marginal queries over $n$ features.
  arXiv  Detail & Related papers  (2024-06-07T12:07:16Z)
• Already Moderate Population Sizes Provably Yield Strong Robustness to Noise [53.27802701790209]
  We show that two evolutionary algorithms can tolerate constant noise probabilities without increasing the runtime on the OneMax benchmark. Our results are based on a novel proof argument that the noiseless offspring can be seen as a biased uniform crossover between the parent and the noisy offspring.
  arXiv  Detail & Related papers  (2024-04-02T16:35:52Z)
• Robust Sparse Estimation for Gaussians with Optimal Error under Huber Contamination [42.526664955704746]
  We study sparse estimation tasks in Huber's contamination model with a focus on mean estimation, PCA, and linear regression. For each of these tasks, we give the first sample and computationally efficient robust estimators with optimal error guarantees. At the technical level, we develop a novel multidimensional filtering method in the sparse regime that may find other applications.
  arXiv  Detail & Related papers  (2024-03-15T15:51:27Z)
• A Sub-Quadratic Time Algorithm for Robust Sparse Mean Estimation [6.853165736531941]
  We study the algorithmic problem of sparse mean estimation in the presence of adversarial outliers. Our main contribution is an algorithm for robust sparse mean estimation which runs in emphsubquadratic time using $mathrmpoly(k,log d,1/epsilon)$ samples.
  arXiv  Detail & Related papers  (2024-03-07T18:23:51Z)
• Error-mitigated fermionic classical shadows on noisy quantum devices [0.3775283002059579]
  Classical shadow (CS) algorithms offer a solution by reducing the number of quantum state copies needed. We propose an error-mitigated CS algorithm assuming gate-independent, time-stationary, and Markovian (GTM) noise. Our algorithm efficiently estimates $k$-RDMs with $widetildemathcal O(knk)$ state copies and $widetildemathcal O(sqrtn)$ calibration measurements for GTM noise.
  arXiv  Detail & Related papers  (2023-10-19T13:27:19Z)
• Near-Optimal Bounds for Learning Gaussian Halfspaces with Random Classification Noise [50.64137465792738]
  We show that any efficient SQ algorithm for the problem requires sample complexity at least $Omega(d1/2/(maxp, epsilon)2)$. Our lower bound suggests that this quadratic dependence on $1/epsilon$ is inherent for efficient algorithms.
  arXiv  Detail & Related papers  (2023-07-13T18:59:28Z)
• Private estimation algorithms for stochastic block models and mixture models [63.07482515700984]
  General tools for designing efficient private estimation algorithms. First efficient $(epsilon, delta)$-differentially private algorithm for both weak recovery and exact recovery.
  arXiv  Detail & Related papers  (2023-01-11T09:12:28Z)
• Robust Sparse Mean Estimation via Sum of Squares [42.526664955704746]
  We study the problem of high-dimensional sparse mean estimation in the presence of an $epsilon$-fraction of adversarial outliers. Our algorithms follow the Sum-of-Squares based, to algorithms approach.
  arXiv  Detail & Related papers  (2022-06-07T16:49:54Z)
• Gaussian Process Bandit Optimization with Few Batches [49.896920704012395]
  We introduce a batch algorithm inspired by finite-arm bandit algorithms. We show that it achieves the cumulative regret upper bound $Oast(sqrtTgamma_T)$ using $O(loglog T)$ batches within time horizon $T$. In addition, we propose a modified version of our algorithm, and characterize how the regret is impacted by the number of batches.
  arXiv  Detail & Related papers  (2021-10-15T00:54:04Z)
• Learning Halfspaces with Tsybakov Noise [50.659479930171585]
  We study the learnability of halfspaces in the presence of Tsybakov noise. We give an algorithm that achieves misclassification error $epsilon$ with respect to the true halfspace.
  arXiv  Detail & Related papers  (2020-06-11T14:25:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.