The Space Complexity of Approximating Logistic Loss

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dexter, Gregory, Drineas, Petros, Khanna, Rajiv
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910725827461120
author Dexter, Gregory
Drineas, Petros
Khanna, Rajiv
author_facet Dexter, Gregory
Drineas, Petros
Khanna, Rajiv
contents We provide space complexity lower bounds for data structures that approximate logistic loss up to $ε$-relative error on a logistic regression problem with data $\mathbf{X} \in \mathbb{R}^{n \times d}$ and labels $\mathbf{y} \in \{-1,1\}^d$. The space complexity of existing coreset constructions depend on a natural complexity measure $μ_\mathbf{y}(\mathbf{X})$, first defined in (Munteanu, 2018). We give an $\tildeΩ(\frac{d}{ε^2})$ space complexity lower bound in the regime $μ_\mathbf{y}(\mathbf{X}) = O(1)$ that shows existing coresets are optimal in this regime up to lower order factors. We also prove a general $\tildeΩ(d\cdot μ_\mathbf{y}(\mathbf{X}))$ space lower bound when $ε$ is constant, showing that the dependency on $μ_\mathbf{y}(\mathbf{X})$ is not an artifact of mergeable coresets. Finally, we refute a prior conjecture that $μ_\mathbf{y}(\mathbf{X})$ is hard to compute by providing an efficient linear programming formulation, and we empirically compare our algorithm to prior approximate methods.
format Preprint
id arxiv_https___arxiv_org_abs_2412_02639
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle The Space Complexity of Approximating Logistic Loss
Dexter, Gregory
Drineas, Petros
Khanna, Rajiv
Data Structures and Algorithms
Machine Learning
We provide space complexity lower bounds for data structures that approximate logistic loss up to $ε$-relative error on a logistic regression problem with data $\mathbf{X} \in \mathbb{R}^{n \times d}$ and labels $\mathbf{y} \in \{-1,1\}^d$. The space complexity of existing coreset constructions depend on a natural complexity measure $μ_\mathbf{y}(\mathbf{X})$, first defined in (Munteanu, 2018). We give an $\tildeΩ(\frac{d}{ε^2})$ space complexity lower bound in the regime $μ_\mathbf{y}(\mathbf{X}) = O(1)$ that shows existing coresets are optimal in this regime up to lower order factors. We also prove a general $\tildeΩ(d\cdot μ_\mathbf{y}(\mathbf{X}))$ space lower bound when $ε$ is constant, showing that the dependency on $μ_\mathbf{y}(\mathbf{X})$ is not an artifact of mergeable coresets. Finally, we refute a prior conjecture that $μ_\mathbf{y}(\mathbf{X})$ is hard to compute by providing an efficient linear programming formulation, and we empirically compare our algorithm to prior approximate methods.
title The Space Complexity of Approximating Logistic Loss
topic Data Structures and Algorithms
Machine Learning
url https://arxiv.org/abs/2412.02639