Bubbly Solutions

We can often employ analog physics to solve computational problems. The earliest example in history is the millstones in windmills from 17th-century Europe. James Watt later improved this concept to regulate and “smooth out” the steam engine. Typically, it demonstrated the following feedback loop:

• If the steam valve outbursts, it would spin the turbine faster

• Then, the balls would swing out as the rotational velocity increases

• It would then close up the valve more and more, reducing steam output

• Consequently, it will slow down the turbine.

Since then, humans have harnessed many simple physical objects, lives, and phenomena to solve “higher-purpose” problems.

Soap Film Computers & Analog Computing

Imagine you need to lay a network cable in a new town using the least length of wires. This is a typical Minimum Spanning Tree (MST) problem, and typically, it will take at least linear time to solve it, even after providing that we already know the distance between each point.

But there is an O(1) way of solving such problems: soap films. If you dip in soap water, the membranes created between each point will always take the shortest path possible due to its surface tension. Like this, using physical and natural phenomena often suggested an extremely efficient or reliable way of optimizing and solving “grand” problems.

• Bird Flocks → Swarm Intelligence. Birds and insects’ collective behavior principles are used in robotics to coordinate multiple robots for search and rescue missions or exploration of unknown environments. Furthermore, stigmergy, indirect coordination between agents by modifying their environments observed in social insects, optimized distributed computing.

• Slime Molds → Network Optimization. This simple organism’s ability to find the shortest path between food sources in a maze has been used to solve network optimization problems, such as finding the most efficient layout for railway networks or optimizing the design of electrical circuits.

• Ant Colony → Pathfinding Optimizations. This algorithm is inspired by ants’ behavior when searching for food. As ants move, they leave behind a pheromone trail that other ants follow. Shorter paths tend to have a higher concentration of pheromones, encouraging more ants to follow them. This principle can be applied to optimize routing problems, such as finding the shortest path in a network.

• DNAs → Molecular Computing. DNA molecules can store and process information due to their ability to self-assemble and the specificity of base pairing. DNA computing has been used to solve complex computational problems, such as the Hamiltonian path problem, which uses finding a path that visits each vertex in a graph exactly once.

• Quantum Fluctuations → Quantum Annealings. Quantum annealing is a method of solving optimization problems using quantum fluctuations. In quantum annealing, a system is initially in a superposition of all possible states and then slowly evolves towards the lowest energy state, corresponding to the optimal solution. This has been used to solve problems such as the traveling salesperson problem and machine learning tasks.