arXiv research

Paper simplifies stock market theory without randomness.

Motivated by the task of hyperparameter optimization, we introduce the non-stochastic best-arm identification problem. Within the multi-armed bandit literature, the cumulative regret objective enjoys algorithms and analyses for both the non-stochastic and stochastic settings while to the best of our knowledge, the best…

Overview of non-stochastic-gradient SA algorithms in signal processing and ML.

This paper makes a small step towards a non-stochastic version of superhedging duality relations in the case of one traded security with a continuous price path. Namely, we prove the coincidence of game-theoretic and measure-theoretic expectation for lower semicontinuous positive functionals. We consider a new broad de…

Study of affine processes without stochastic continuity assumption.

New controller reduces regret in non-stochastic control with adversarial perturbations.

New algorithm achieves optimal regret in non-stochastic control, showing stochasticity is not beneficial.

This paper establishes a non-stochastic analogue of the celebrated result by Dubins and Schwarz about reduction of continuous martingales to Brownian motion via time change. We consider an idealized financial security with continuous price path, without making any stochastic assumptions. It is shown that typical price …

Study finds optimal regret bound for multi-armed bandit problem with expert advice.

New algorithm minimizes worst-case regret in uncertain, time-varying dynamics.

Improved MCTS for asymmetric trees and loops.

We consider a non-stochastic online learning approach to price financial options by modeling the market dynamic as a repeated game between the nature (adversary) and the investor. We demonstrate that such framework yields analogous structure as the Black-Scholes model, the widely popular option pricing model in stochas…

New algorithm reduces decision switching in dynamic environments.

New algorithm reduces regret in multi-player bandits with collision information.

Forré introduces a new conditional independence notion for mixed variables.

This work addresses the problem of regret minimization in non-stochastic multi-armed bandit problems, focusing on performance guarantees that hold with high probability. Such results are rather scarce in the literature since proving them requires a large deal of technical effort and significant modifications to the sta…

Develops a new framework for conditional independence.

Guaranteed bounds for posterior inference in probabilistic programs.

Study optimal arms in combinatorial bandits with semi-bandit feedback and finite budget.

New method tracks significant arm switches to improve bandit algorithms.

We propose a framework to perform streaming covariance selection. Our approach employs regularization constraints where a time-varying sparsity parameter is iteratively estimated via stochastic gradient descent. This allows for the regularization parameter to be efficiently learnt in an online manner. The proposed fram…

New algorithm controls systems with unknown, changing losses.

We design differentially private algorithms for the problem of online linear optimization in the full information and bandit settings with optimal $\tilde{O}(\sqrt{T})$ regret bounds. In the full-information setting, our results demonstrate that $ε$-differential privacy may be ensured for free -- in particular, the reg…

Improved privacy analysis for stochastic gradient descent.

New algorithm controls linear systems with bandit feedback, achieving optimal regret.

Performance of machine learning algorithms depends critically on identifying a good set of hyperparameters. While recent approaches use Bayesian optimization to adaptively select configurations, we focus on speeding up random search through adaptive resource allocation and early-stopping. We formulate hyperparameter op…

New model for display advertising with stochastic and adversarial components.

We consider derivative-free algorithms for stochastic and non-stochastic convex optimization problems that use only function values rather than gradients. Focusing on non-asymptotic bounds on convergence rates, we show that if pairs of function values are available, algorithms for $d$-dimensional optimization that use …

The problem of non-stationarity in financial markets is discussed and related to the dynamic nature of price volatility. A new measure is proposed for estimation of the current asset volatility. A simple and illustrative explanation is suggested of the emergence of significant serial autocorrelations in volatility and …

We demonstrate that, in the classical non-stochastic regret minimization problem with $d$ decisions, gains and losses to be respectively maximized or minimized are fundamentally different. Indeed, by considering the additional sparsity assumption (at each stage, at most $s$ decisions incur a nonzero outcome), we derive…

Optimal trading patterns adjust based on market efficiency and slippage costs.

Stochastic gradient descent is the method of choice for large-scale machine learning problems, by virtue of its light complexity per iteration. However, it lags behind its non-stochastic counterparts with respect to the convergence rate, due to high variance introduced by the stochastic updates. The popular Stochastic …

Unified analysis of FL with arbitrary client participation.

DISCERN estimates K for K-Means clustering efficiently.

This paper proposes a self-supervised exploration method using disagreement of dynamics models.

Unified framework for analyzing online convex optimization across various settings.

A new federated bandit problem with multiple adversaries, solved with a near-optimal algorithm.

MetaGrad adapts multiple learning rates for faster online optimization.

AdaGrad converges to stationary points in nonconvex optimization with robust stepsize.

Learning theory has largely focused on two main learning scenarios. The first is the classical statistical setting where instances are drawn i.i.d. from a fixed distribution and the second scenario is the online learning, completely adversarial scenario where adversary at every time step picks the worst instance to pro…

Paper develops bandit algorithms for nonstationary nonconvex optimization.

A heuristic method for determining input ranges for complex processes.

We consider high dimensional sparse regression, and develop strategies able to deal with arbitrary -- possibly, severe or coordinated -- errors in the covariance matrix $X$. These may come from corrupted data, persistent experimental errors, or malicious respondents in surveys/recommender systems, etc. Such non-stochas…

SGD vs quasi-Newton optimization in neural networks: different landscapes, different generalizability.

The paper studies continuous submodular functions and their optimization.

Solves complex Monge-Ampère equation with Hölder continuous boundary data.

Bilevel Continual Learning improves continual learning by transferring knowledge effectively.

Uniform Lipschitz continuity of isoperimetric profiles in evolving surfaces.