arXiv research

Two new algorithms solve nonconvex-strongly concave problems efficiently.

A novel decentralized algorithm improves minimax optimization in federated learning.

We investigate the use of Minimax distances to extract in a nonparametric way the features that capture the unknown underlying patterns and structures in the data. We develop a general-purpose and computationally efficient framework to employ Minimax distances with many machine learning methods that perform on numerica…

Many tasks in modern machine learning can be formulated as finding equilibria in \emph{sequential} games. In particular, two-player zero-sum sequential games, also known as minimax optimization, have received growing interest. It is tempting to apply gradient descent to solve minimax optimization given its popularity a…

Paper explores generalization of minimax learners, proposing a new metric.

A new method solves complex constrained minimax problems.

Convergence to a saddle point for convex-concave functions has been studied for decades, while recent years has seen a surge of interest in non-convex (zero-sum) smooth games, motivated by their recent wide applications. It remains an intriguing research challenge how local optimal points are defined and which algorith…

New algorithm solves federated minimax optimization problems.

Paper proposes an algorithm to solve complex minimax problems efficiently.

A new decentralized method solves minimax problems with reduced communication and sample complexity.

Efficient sampling reduces memory usage for Minimax distance analysis.

Study on estimating invertible functions with minimax analysis.

Minimax linkage was first introduced by Ao et al. [3] in 2004, as an alternative to standard linkage methods used in hierarchical clustering. Minimax linkage relies on distances to a prototype for each cluster; this prototype can be thought of as a representative object in the cluster, hence improving the interpretabil…

New bounds on minimax regret for sequential probability assignment using logarithmic loss.

Paper proposes algorithms for solving nonconvex-nonconcave problems with complexity guarantees.

New algorithm AG-OG optimizes separable convex-concave problems efficiently.

Develops shuffling gradient-based methods for nonconvex-concave minimax optimization.

Paper improves risk bounds for nonconvex-strongly-concave minimax problems.

Optimally estimate distances on surfaces using reconstructed meshes.

Study minimax linear regression under quantile risk, improving existing bounds and providing new results.

The paper analyzes generalization bounds for NC-SC/NC-C stochastic minimax optimization.

This paper analyzes how machine learning models resist adversarial attacks in nonparametric regression.

Develops high-probability minimax quantile bounds for statistical problems.

TiAda adapts adaptive gradient methods for nonconvex minimax optimization.

New algorithms solve nonconvex-nonconcave minimax optimization problems.

New estimator achieves minimax optimal risk in transfer learning.

We prove a new minimax theorem connecting the worst-case Bayesian regret and minimax regret under partial monitoring with no assumptions on the space of signals or decisions of the adversary. We then generalise the information-theoretic tools of Russo and Van Roy (2016) for proving Bayesian regret bounds and combine th…

Minimax solutions are weak solutions to Cauchy problems involving Hamilton--Jacobi equations, constructed from generating families quadratic at infinity of their geometric solutions. We give a complete description of minimax solutions and we classify their generic singularities of codimension not greater than 2.

This paper tackles convex-submodular minimax problems in mixed continuous-discrete domains.

Paper proves Sion's theorem in geodesic spaces and develops a Riemannian extragradient method.

Adversarial meta-learning computes Gamma-minimax estimators for vague prior knowledge.

The paper analyzes how optimization algorithms affect the generalization of minimax models.

New analysis improves understanding of bilevel optimization stability and generalization.

New method solves complex constrained optimization problems.

The paper optimizes k-NN for distributed learning with minimax optimal performance.

Since its inception, the modus operandi of multi-task learning (MTL) has been to minimize the task-wise mean of the empirical risks. We introduce a generalized loss-compositional paradigm for MTL that includes a spectrum of formulations as a subfamily. One endpoint of this spectrum is minimax MTL: a new MTL formulation…

State-of-the-art adversarial attacks are aimed at neural network classifiers. By default, neural networks use gradient descent to minimize their loss function. The gradient of a classifier's loss function is used by gradient-based adversarial attacks to generate adversarially perturbed images. We pose the question whet…

We study the problem of switching-constrained online convex optimization (OCO), where the player has a limited number of opportunities to change her action. While the discrete analog of this online learning task has been studied extensively, previous work in the continuous setting has neither established the minimax ra…

The paper develops a new algorithm for constructing minimax estimators using online learning techniques.

This paper studies continuum-armed bandits under Besov smoothness conditions and derives minimax rates.

A new method solves a complex optimization problem efficiently.

Study minimax optimal RL in factored MDPs with bonus exploration.

In this paper we give sufficient conditions guaranteeing the validity of the well-known minimax theorem for the lower Snell envelope with respect to a family of absolutely continuous probability measures. Such minimax results play an important role in the characterisation of arbitrage-free prices of American contingent…

This paper explores minimax-Bayes solutions for reinforcement learning problems.

Simple proof shows forecasts can be calibrated in a few periods.

Classical scaling is shown to be optimal under various noisy conditions.

Study exact minimax rates for density estimation over convex classes, extending previous work.

Choquet and minimax expectations are equivalent in European option pricing.