arXiv Open Access 2015

How to show a probabilistic model is better

Mithun Chakraborty Sanmay Das Allen Lavoie

Lihat Sumber

Abstrak

We present a simple theoretical framework, and corresponding practical procedures, for comparing probabilistic models on real data in a traditional machine learning setting. This framework is based on the theory of proper scoring rules, but requires only basic algebra and probability theory to understand and verify. The theoretical concepts presented are well-studied, primarily in the statistics literature. The goal of this paper is to advocate their wider adoption for performance evaluation in empirical machine learning.

Topik & Kata Kunci

stat.ML cs.LG

Penulis (3)

Mithun Chakraborty

Sanmay Das

Allen Lavoie

Format Sitasi

APA MLA BibTeX

Chakraborty, M., Das, S., Lavoie, A. (2015). How to show a probabilistic model is better. https://arxiv.org/abs/1502.03491

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2015
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓