This talk was given at Metis (San Francisco) on October 11, 2017. The livestream can be viewed here: https://livestream.com/metis/events/7706066

This material builds on what you already know about parameter tuning to dive deeper into the world of stochastic optimization. In particular, we’ll explore a range of methods, including stochastic gradient descent, simulated annealing, and particle swarm optimization, and cultivated intuition on how these approaches work, when they can be applied, what their underlying topologies look like, and how you can get the best performance out of them.

Stochastic Multi-Armed Bandits

New approaches for solving the Probabilistic Traveling Salesman Problem

Sklearn thoughts on tuning

A Practical Guide to Support Vector Classification

Support Vector Machines for Business Applications

Statistical Performance of Support Vector Machines

Breaking the Curse of Kernelization: Budgeted Stochastic Gradient Descent for Large-Scale SVM Training

A stochastic optimization approach for parameter tuning of Support Vector Machines

SVM Parameters Tuning with Quantum Particles Swarm Optimization

Particle swarm optimization for parameter determination and feature selection of support vector machines

Accelerated Particle Swarm Optimization and Support Vector Machine for Business Optimization and Applications

Perspective: Energy Landscapes for Machine Learning

Efficient Tuning of SVM Hyperparameters Using Radius/Margin Bound and Iterative Algorithms