Intuition of particle swarm optimization
source link: https://dev.to/rishalhurbans/intuition-of-particle-swarm-optimization-32m4
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
Intuition of particle swarm optimization
Swarm intelligence is an amazing phenomena in nature. We see it in flocks of birds, bees in a hive, bacterial growth, and more. The behaviour of these wonderful creatures have been studied and inspired useful algorithms. Here's an introduction to particle swarm optimization.
Particle swarm optimization is a swarm intelligence algorithm. It relies on emergent behavior of many individuals to solve difficult problems as a collective. The most amazing feats are always achieved as a collective. Learn more about swarms here: https://rhurbans.com/optimization-finding-the-best-solutions/
Bird flocks are an ideal example of swarm intelligence in nature. When a single bird is flying, it might attempt several maneuvers and techniques to preserve energy - jumping and gliding or leveraging wind currents to carry it in the right direction of travel.
This behavior indicates some primitive level of intelligence in a single individual. But birds also have the need to migrate during different seasons. In winter, there is less availability of insects and food, nesting locations are scarce, and warm weather is better for survival.
Migration is usually not a short trip. It takes thousands of kilometers of movement to arrive at an area with suitable conditions. When birds travel these long distances, they tend to flock.
The formation that we observe in bird flocks has several advantages. A large, strong bird will take the lead, and when it flaps its wings, it creates uplift for the birds behind it. These birds can fly while using significantly less energy.
Particle swarm optimization involves a group of individuals at different points in the solution space, all using real-life swarm concepts to find an optimal solution. Imagine building a drone and finding the optimal plastic, aluminium, while minimizing drag in the wind.
Each combination of the values for plastic, aliminium, and drag is a possible solution. We want to find the values for plastic and aluminium that minimize the drag value. Particle swarm optimization can help us with this.
PSOs are useful in large search spaces —There are many data points and possibilities of combinations. And search spaces with high dimensions —There are many parameters that influence how well a solution performs.
A PSO particle contains a:
- Position —The position of the particle in all dimensions
- Best position —The best position found using the fitness function
- Velocity —The current velocity of the particle’s movement
Particle swarm optimization provides a means to search a large search space without checking every value in each dimension. These attributes help particles search the solution space and converge to good solutions inspired by swarms in nature.
If you found this interesting and useful, see Grokking AI Algorithms with Manning Publications: http://bit.ly/gaia-book, consider following me - @RishalHurbans, or join my mailing list for infrequent knowledge drops: https://rhurbans.com/subscribe.
Recommend
-
42
-
46
-
15
Wondering why deep learning works? Here is the intuition behind it. This blog is wholly inspired by the lectures of Professor...
-
33
In the physical world, nanotechnology predictably brings us surprises. Being structured on nano-scale, the most common materials, such as carbon, iron or copper, gain new unforeseen properties. What we know about them on t...
-
21
An imbalanced classification problem is a problem that involves predicting a class label where the distribution of class labels in the training dataset is not equal. A challenge for beginners working with imbalan...
-
16
Risk-takers | 100x engineers | Astronauts | From Unsplash Y ou will always hear this and it’s also in...
-
13
Dataset is published for explaining the capability of each applicant’s repaying a loan? Below is the distribution of the Target feature and some of the independent features. Target feature has imbalanced data problem becau...
-
7
What's the point of pattern matching if we already have conditionals and variable assignment in a language? Pattern matching helps tease apart values and construct control flow using the shape of data rather than b...
-
11
Important inequalities in convex optimization, proofs and intuition Many talk about data science and machine learning with enthusiasm, but few know about one of the most important building components behind them – con...
-
9
Autonomous Navigation with Brian Douglas: Part 5 This post is from Brian Douglas, YouTube content creator for Control Systems and Autonomous Applications.
About Joyk
Aggregate valuable and interesting links.
Joyk means Joy of geeK