Convex optimization over a probability simplex

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chok, James, Vasil, Geoffrey M.
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910959263547392
author Chok, James
Vasil, Geoffrey M.
author_facet Chok, James
Vasil, Geoffrey M.
contents We propose a new iteration scheme, the Cauchy-Simplex, to optimize convex problems over the probability simplex $\{w\in\mathbb{R}^n\ |\ \sum_i w_i=1\ \textrm{and}\ w_i\geq0\}$. Specifically, we map the simplex to the positive quadrant of a unit sphere, envisage gradient descent in latent variables, and map the result back in a way that only depends on the simplex variable. Moreover, proving rigorous convergence results in this formulation leads inherently to tools from information theory (e.g., cross-entropy and KL divergence). Each iteration of the Cauchy-Simplex consists of simple operations, making it well-suited for high-dimensional problems. In continuous time, we prove that $f(x_T)-f(x^*) = {O}(1/T)$ for differentiable real-valued convex functions, where $T$ is the number of time steps and $w^*$ is the optimal solution. Numerical experiments of projection onto convex hulls show faster convergence than similar algorithms. Finally, we apply our algorithm to online learning problems and prove the convergence of the average regret for (1) Prediction with expert advice and (2) Universal Portfolios.
format Preprint
id arxiv_https___arxiv_org_abs_2305_09046
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Convex optimization over a probability simplex
Chok, James
Vasil, Geoffrey M.
Optimization and Control
Machine Learning
Numerical Analysis
Portfolio Management
65K10, 68W27, 68W40, 91G10, 97U40
We propose a new iteration scheme, the Cauchy-Simplex, to optimize convex problems over the probability simplex $\{w\in\mathbb{R}^n\ |\ \sum_i w_i=1\ \textrm{and}\ w_i\geq0\}$. Specifically, we map the simplex to the positive quadrant of a unit sphere, envisage gradient descent in latent variables, and map the result back in a way that only depends on the simplex variable. Moreover, proving rigorous convergence results in this formulation leads inherently to tools from information theory (e.g., cross-entropy and KL divergence). Each iteration of the Cauchy-Simplex consists of simple operations, making it well-suited for high-dimensional problems. In continuous time, we prove that $f(x_T)-f(x^*) = {O}(1/T)$ for differentiable real-valued convex functions, where $T$ is the number of time steps and $w^*$ is the optimal solution. Numerical experiments of projection onto convex hulls show faster convergence than similar algorithms. Finally, we apply our algorithm to online learning problems and prove the convergence of the average regret for (1) Prediction with expert advice and (2) Universal Portfolios.
title Convex optimization over a probability simplex
topic Optimization and Control
Machine Learning
Numerical Analysis
Portfolio Management
65K10, 68W27, 68W40, 91G10, 97U40
url https://arxiv.org/abs/2305.09046