Hierarchical Density Factorization with Chocloton
Density Estimation using Multi-Agent Optimization & Rewards
Introduction
The purpose, problem statement, and potential applications came from this post on datasciencecentral.com. The goal is to approximate any multi-variate distribution using a weighted sum of kernels. Here, a kernel refers to a parameterized distribution. This method of using a decaying weighted sum of kernels to approximate a distribution is similar to a Taylor series where a function can be approximated, around a point, using the function’s derivatives.
Goals
- Approximate any empirical distribution
- Build a parameterized density estimator
- Outlier detection and dataset noise reduction
The Approach
This solution I came up with was incorporated into a python package, Chocloton. The example code can be found here.
My solution uses the following:
1. Particle Swarm\Genetic Optimizer
2. Multi-Agent Approximation using IID Kernels
3. Reinforcement Learning
Use Cases
- With Density Factorization you can get multivariate density estimate, cluster assignment, and similar data points all in one shot.
- After training, the HDRE model can assign a matrix of…