Swarm AI



Groups are smarter when they think together. Individuals would be on their own. Natural species like birds, bees, fish, ants and many other organisms work together as an organized system to hunt, build homes and survive as a pack. Scientists call the swarm intelligence which is a combining many minds into one. People can swarm now with the use of artificial swarm intelligence, we can quickly and accurately respond to questions, reach decisions, and even making predictions, generate forecasts, optimize opinions. A combination of AI algorithms and human input to amplify collective intelligence when solving problems and making informed decisions.

How does it work?

Humans didn’t evolve the in-born ability to form subtle connections that natural species use to communicate between one another. However, with the help of blazingly high speed networking technology, we can communicate with each other from anywhere without the geographical and time limitations. What we need to achieve is to create a real-time system with instant feedbacks within a closed community.

The 21st Century is a digital book. Every person, every organization and every individual entity leaves footprints and traces on the internet. Exchange of information, knowledge allows many of us to contribute to the rich contents of the digital book, and we can have a great understanding of human behaviour using the statistical data we collected in the virtual world, to improve the society.

From forecasting financial trends and presidential elections to assessing the effectiveness of advertisements. While the performance of machine learning models has proven that we have great rooms for improvement of the accuracy, precision and recall of the models, it shows great potential of massive application of swarm intelligence. It’s easy for us to gather evidence relevant for analysing situations, but the data itself is disorganized and ‘noisy’. Therefore, a novel interface can allow participants to convey their individual intent in parallel, letting users perceive and react to the changing system in real-time, and close a feedback loop around the full population of the swarm community.

I recently came across with hardcore gamers using Swarm AI techniques in games and 3D virtual environment to experiment on players stuck and escaping a tiny cave. AI look around and find the exit map the way out. This functionality enabled players to place beacons inside hidden caves that the AI will be able to navigate to and setup “ambushes” as well as giving the AI the ability to “hide” and get out of hiding places smoothly within the game.  The demo shows great possibilities of how swarm AI could also be implemented in automated rescue missions where places are within human reaches or with great threats when drone swarms powered by AI can complete the tasks men can’t achieve.

I am not an expert of swarm AI but I truly believe that swarm AI gives us a great direction that collective intelligence is the pathway of the future. Decades from now, humans will be able to increase their efficiency using high speed networks to collaborate with each other effectively to solve problems from different locations around the world. Laying the groundwork of the network infrastructure would be our main task for the next 10 years as well as how the network work. Humans often demonstrates its position as a dominant species in nature, yet we still learn a lot more from species that are considered not as smart as we do. When technological advancement reaches a certain level of greatness, it will eventually fall back to connecting with nature, given the fact that we are part of the nature afterall.