Overview / Usage

Imagine you're working with a dataset with 10 features. Using all the 10 features, you're able to make accurate prediction 95% of the times.

Now, let's assume your training algorithm has an N^3 complexity.

Thus, to get a 95% accuracy, you'd be taking 1000 units of time to train.

Assuming that after feature selection, you are able to drop 6 features, and are able to still have a 95% accuracy with just 4 features, your training time comes down to 64 units of time.

When I do the math, I love just spending 64 units of time (and computational effort) compared to 1000 units.

​

This library introduces binary swarm algorithms which have been used to perform feature selection (could easily be used for other tasks as well!)

Methodology / Approach

Swarm optimizers are a part of the **Heuristic Algorithms** family. A heuristic algorithm is one that is designed to solve a problem in a faster and more efficient fashion than traditional methods by sacrificing optimality, accuracy, precision, or completeness for speed. Heuristic algorithms often times used to solve NP-complete problems, a class of decision problems.

​

Swarm Intelligence systems employ large numbers of agents interacting locally with one another and the environment. Swarm intelligence refers to the collective behavior of decentralized systems and can be used to describe both natural and artificial systems. Specific algorithms for this class of system include the particle swarm optimization algorithm, the ant colony optimization algorithm, and artificial bee colony algorithm. Each of the previous algorithms was inspired by the natural, self-organized behavior of animals.

Technologies Used

OneAPI Intel Python

DevCloud

Repository

https://github.com/xprilion/binary_swarm_optimization