The Power of Sampling: Dimension-free Risk Bounds in Private ERM

• URL: http://arxiv.org/abs/2105.13637v4
• Date: Mon, 3 Jun 2024 21:31:18 GMT
• Title: The Power of Sampling: Dimension-free Risk Bounds in Private ERM
• Authors: Yin Tat Lee, Daogao Liu, Zhou Lu,
• Abstract summary: We show that our algorithm can achieve rank-dependent risk bounds for non-smooth convex objectives using only zeroth order oracles. This highlights the power of sampling in differential privacy.
• Score: 25.676350220943274
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Differentially private empirical risk minimization (DP-ERM) is a fundamental problem in private optimization. While the theory of DP-ERM is well-studied, as large-scale models become prevalent, traditional DP-ERM methods face new challenges, including (1) the prohibitive dependence on the ambient dimension, (2) the highly non-smooth objective functions, (3) costly first-order gradient oracles. Such challenges demand rethinking existing DP-ERM methodologies. In this work, we show that the regularized exponential mechanism combined with existing samplers can address these challenges altogether: under the standard unconstrained domain and low-rank gradients assumptions, our algorithm can achieve rank-dependent risk bounds for non-smooth convex objectives using only zeroth order oracles, which was not accomplished by prior methods. This highlights the power of sampling in differential privacy. We further construct lower bounds, demonstrating that when gradients are full-rank, there is no separation between the constrained and unconstrained settings. Our lower bound is derived from a general black-box reduction from unconstrained to the constrained domain and an improved lower bound in the constrained setting, which might be of independent interest.

Related papers

• Enlarging Feature Support Overlap for Domain Generalization [9.227839292188346]
  Invariant risk minimization (IRM) addresses this issue by learning invariant features and minimizing the risk across different domains. We propose a novel method to enlarge feature support overlap for domain generalization. Specifically, we introduce Bayesian random data augmentation to increase sample diversity and overcome the deficiency of IRM.
  arXiv  Detail & Related papers  (2024-07-08T09:16:42Z)
• Double Duality: Variational Primal-Dual Policy Optimization for Constrained Reinforcement Learning [132.7040981721302]
  We study the Constrained Convex Decision Process (MDP), where the goal is to minimize a convex functional of the visitation measure. Design algorithms for a constrained convex MDP faces several challenges, including handling the large state space.
  arXiv  Detail & Related papers  (2024-02-16T16:35:18Z)
• Riemannian stochastic optimization methods avoid strict saddle points [68.80251170757647]
  We show that policies under study avoid strict saddle points / submanifolds with probability 1. This result provides an important sanity check as it shows that, almost always, the limit state of an algorithm can only be a local minimizer.
  arXiv  Detail & Related papers  (2023-11-04T11:12:24Z)
• Beyond the Edge of Stability via Two-step Gradient Updates [49.03389279816152]
  Gradient Descent (GD) is a powerful workhorse of modern machine learning. GD's ability to find local minimisers is only guaranteed for losses with Lipschitz gradients. This work focuses on simple, yet representative, learning problems via analysis of two-step gradient updates.
  arXiv  Detail & Related papers  (2022-06-08T21:32:50Z)
• Convergence and sample complexity of natural policy gradient primal-dual methods for constrained MDPs [21.347689976296834]
  We employ the natural policy gradient method to solve the discounted optimal optimal rate problem. We also provide convergence and finite-sample guarantees for two sample-based NPG-PD algorithms.
  arXiv  Detail & Related papers  (2022-06-06T04:28:04Z)
• Improving Generalization via Uncertainty Driven Perturbations [107.45752065285821]
  We consider uncertainty-driven perturbations of the training data points. Unlike loss-driven perturbations, uncertainty-guided perturbations do not cross the decision boundary. We show that UDP is guaranteed to achieve the robustness margin decision on linear models.
  arXiv  Detail & Related papers  (2022-02-11T16:22:08Z)
• Keep it Tighter -- A Story on Analytical Mean Embeddings [0.6445605125467574]
  Kernel techniques are among the most popular and flexible approaches in data science. Mean embedding gives rise to a divergence measure referred to as maximum mean discrepancy (MMD) In this paper we focus on the problem of MMD estimation when the mean embedding of one of the underlying distributions is available analytically.
  arXiv  Detail & Related papers  (2021-10-15T21:29:27Z)
• A Fully Problem-Dependent Regret Lower Bound for Finite-Horizon MDPs [117.82903457289584]
  We derive a novel problem-dependent lower-bound for regret in finite-horizon Markov Decision Processes (MDPs) We show that our lower-bound is considerably smaller than in the general case and it does not scale with the minimum action gap at all. We show that this last result is attainable (up to $poly(H)$ terms, where $H$ is the horizon) by providing a regret upper-bound based on policy gaps for an optimistic algorithm.
  arXiv  Detail & Related papers  (2021-06-24T13:46:09Z)
• On Lower Bounds for Standard and Robust Gaussian Process Bandit Optimization [55.937424268654645]
  We consider algorithm-independent lower bounds for the problem of black-box optimization of functions having a bounded norm. We provide a novel proof technique for deriving lower bounds on the regret, with benefits including simplicity, versatility, and an improved dependence on the error probability.
  arXiv  Detail & Related papers  (2020-08-20T03:48:14Z)
• Continuous Profit Maximization: A Study of Unconstrained Dr-submodular Maximization [4.649999862713523]
  We form continuous profit (CPM-MS) problem, whose domain is on integer lattices. We introduce lattice-based double greedy algorithm, which can obtain a constant approximation. We propose a technique, called lattice-based iterative pruning. It can shrink the search space effectively.
  arXiv  Detail & Related papers  (2020-04-12T05:35:19Z)

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.