Jensen: An Easily-Extensible C++ Toolkit for Production-Level Machine Learning and Convex Optimization

TL;DR

Jensen is a scalable, extensible C++ toolkit designed for efficient production-level machine learning and convex optimization, enabling easy deployment and customization of models and algorithms.

Contribution

It introduces a flexible framework that integrates various convex functions, optimization algorithms, and machine learning models, facilitating rapid development and deployment.

Findings

Supports multiple optimization algorithms including Gradient Descent and L-BFGS

Enables deployment of models with minimal code

Allows easy extension with new loss functions and algorithms

Abstract

This paper introduces Jensen, an easily extensible and scalable toolkit for production-level machine learning and convex optimization. Jensen implements a framework of convex (or loss) functions, convex optimization algorithms (including Gradient Descent, L-BFGS, Stochastic Gradient Descent, Conjugate Gradient, etc.), and a family of machine learning classifiers and regressors (Logistic Regression, SVMs, Least Square Regression, etc.). This framework makes it possible to deploy and train models with a few lines of code, and also extend and build upon this by integrating new loss functions and optimization algorithms.