A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Zeroth-order optimization methods lack inherent privacy guarantees.
New method for zeroth-order stochastic gradient algorithms provides confidence intervals.
In this paper, we propose and analyze zeroth-order stochastic approximation algorithms for nonconvex and convex optimization, with a focus on addressing constrained optimization, high-dimensional setting and saddle-point avoiding. To handle constrained optimization, we first propose generalizations of the conditional g…
A new method for MARL with partial observations reduces communication overhead.
Discretizations of Langevin diffusions provide a powerful method for sampling and Bayesian inference. However, such discretizations require evaluation of the gradient of the potential function. In several real-world scenarios, obtaining gradient evaluations might either be computationally expensive, or simply impossibl…
This paper reviews zeroth-order optimization in signal processing and machine learning.
We consider the problem of optimizing a high-dimensional convex function using stochastic zeroth-order queries. Under sparsity assumptions on the gradients or function values, we present two algorithms: a successive component/feature selection algorithm and a noisy mirror descent algorithm using Lasso gradient estimate…
LAZO reduces query complexity and variance in ZO methods.
This paper analyzes and guarantees convergence of prior-guided ZO algorithms.
Proximal gradient method has been playing an important role to solve many machine learning tasks, especially for the nonsmooth problems. However, in some machine learning problems such as the bandit model and the black-box learning problem, proximal gradient method could fail because the explicit gradients of these pro…
New method improves zeroth-order stochastic optimization with adaptive sampling.
A new method for optimizing functions without gradients, improving efficiency and convergence.
New algorithms solve complex minimax problems without needing derivatives.
We propose a method for zeroth order stochastic convex optimization that attains the suboptimality rate of after queries for a convex bounded function . The method is based on a random walk (the \emph{Ball Walk}) on the epigraph of the function. Th…
ConMeZO speeds up zeroth-order optimization for large language models.
ZDPG learns model-free policies without critics, improving on PG.
ZOSPI improves RL policies with global value function exploitation.
As application demands for zeroth-order (gradient-free) optimization accelerate, the need for variance reduced and faster converging approaches is also intensifying. This paper addresses these challenges by presenting: a) a comprehensive theoretical analysis of variance reduced zeroth-order (ZO) optimization, b) a nove…
Zeroth-order methods favor flat minima in machine learning.
Alternating direction method of multipliers (ADMM) is a popular optimization tool for the composite and constrained problems in machine learning. However, in many machine learning problems such as black-box attacks and bandit feedback, ADMM could fail because the explicit gradients of these problems are difficult or in…
Despite the great achievements of the modern deep neural networks (DNNs), the vulnerability/robustness of state-of-the-art DNNs raises security concerns in many application domains requiring high reliability. Various adversarial attacks are proposed to sabotage the learning performance of DNN models. Among those, the b…
Improves zeroth-order optimization for private machine learning with public data.
LORENZA improves LLM fine-tuning efficiency and generalization.
New method estimates Riemannian derivatives from noisy function evaluations.
MAZE attacks a model without data, using synthetic inputs.
Two algorithms solve nonconvex minimax problems with linear constraints, achieving complexity guarantees.
In the learning to learn (L2L) framework, we cast the design of optimization algorithms as a machine learning problem and use deep neural networks to learn the update rules. In this paper, we extend the L2L framework to zeroth-order (ZO) optimization setting, where no explicit gradient information is available. Our lea…
Book covers tools for zeroth-order convex optimisation.
Optimal algorithms for Riemannian optimization with reduced complexity.
Paper proposes algorithms for solving nonconvex-nonconcave problems with complexity guarantees.
Paper develops methods to optimize policies directly from human feedback without reward inference.
Paper tackles gradient-free minimax optimization with variance reduction for faster convergence.
New algorithm reduces dimensionality in stochastic optimization.
A new hybrid-ordered SGD method reduces communication and complexity for non-convex optimization.
In this paper, we design and analyze a new zeroth-order online algorithm, namely, the zeroth-order online alternating direction method of multipliers (ZOO-ADMM), which enjoys dual advantages of being gradient-free operation and employing the ADMM to accommodate complex structured regularizers. Compared to the first-ord…
In this paper, we study zeroth-order algorithms for minimax optimization problems that are nonconvex in one variable and strongly-concave in the other variable. Such minimax optimization problems have attracted significant attention lately due to their applications in modern machine learning tasks. We first consider a …
Gradient descent and its variants are widely used in machine learning. However, oracle access of gradient may not be available in many applications, limiting the direct use of gradient descent. This paper proposes a method of estimating gradient to perform gradient descent, that converges to a stationary point for gene…
New optimization method improves generalization across various tasks.
Two types of zeroth-order stochastic algorithms have recently been designed for nonconvex optimization respectively based on the first-order techniques SVRG and SARAH/SPIDER. This paper addresses several important issues that are still open in these methods. First, all existing SVRG-type zeroth-order algorithms suffer …
We consider the problem of global optimization of an unknown non-convex smooth function with zeroth-order feedback. In this setup, an algorithm is allowed to adaptively query the underlying function at different locations and receives noisy evaluations of function values at the queried points (i.e. the algorithm has ac…
In this paper, we propose a new technique named \textit{Stochastic Path-Integrated Differential EstimatoR} (SPIDER), which can be used to track many deterministic quantities of interest with significantly reduced computational cost. We apply SPIDER to two tasks, namely the stochastic first-order and zeroth-order method…
Sparse perturbations improve convergence in SZO methods for faster training.
Zeroth-order optimization is an important research topic in machine learning. In recent years, it has become a key tool in black-box adversarial attack to neural network based image classifiers. However, existing zeroth-order optimization algorithms rarely extract second-order information of the model function. In this…
New Hessian estimators for Riemannian manifolds with reduced bias.
New adaptive methods solve weakly convex stochastic optimization problems.
New method uses zeroth-order queries to approximate proximal sampling efficiently.
DPZero fine-tunes large models privately without backpropagation.
Derivative-free optimization has become an important technique used in machine learning for optimizing black-box models. To conduct updates without explicitly computing gradient, most current approaches iteratively sample a random search direction from Gaussian distribution and compute the estimated gradient along that…