Hamiltonian Monte Carlo Particle Swarm Optimizer
This addresses optimization challenges in machine learning, particularly for non-convex functions and DNNs, but appears incremental as it hybridizes existing methods.
The paper tackles optimization problems by introducing the Hamiltonian Monte Carlo Particle Swarm Optimizer (HMC-PSO), which combines Exponentially Averaged Momentum PSO with HMC sampling to explore non-convex functions efficiently, and extends it to approximate error gradients for Deep Neural Networks.
We introduce the Hamiltonian Monte Carlo Particle Swarm Optimizer (HMC-PSO), an optimization algorithm that reaps the benefits of both Exponentially Averaged Momentum PSO and HMC sampling. The coupling of the position and velocity of each particle with Hamiltonian dynamics in the simulation allows for extensive freedom for exploration and exploitation of the search space. It also provides an excellent technique to explore highly non-convex functions while ensuring efficient sampling. We extend the method to approximate error gradients in closed form for Deep Neural Network (DNN) settings. We discuss possible methods of coupling and compare its performance to that of state-of-the-art optimizers on the Golomb's Ruler problem and Classification tasks.