What Is Heuristics?

Heuristics are like a carpenter's toolkit. The problem is your board, and the solution is the piece of wood you need to cut out. Your toolkit is full of saws and chisels, pliers and mallets, and all made to help you get the job done quickly—even if it isn't perfect. Heuristics is a way of approaching problem-solving that doesn't try to find an ideal solution but instead looks for one that works well enough within a reasonable time frame. Instead of taking as long as possible over every detail, heuristic strategies help us find solutions within an acceptable range of accuracy. Heuristics are a helpful shortcut when trying to solve an impossible problem. Heuristics are all about getting the job done, no matter what. They're like wild cards in poker; they're not ideal, but if you play your hand right, they can turn out okay. In machine learning and artificial intelligence, heuristics are used when solving a particular problem with a step-by-step algorithm is impractical. Because a heuristic approach emphasizes speed over accuracy, it is often combined with optimization algorithms to improve results. Successive iterations are interdependent, and each level of a deep neural network decides which avenues to choose, and discard based on their proximity to the desired solution. Heuristics are the best way to solve small-scale problems when you don't have the time or resources to solve the big problem. A heuristic is a rule of thumb that quickly but inaccurately solves a problem. It's often used when insufficient time or money is available to find an exact solution. Heuristics are used by humans and machines alike—they're basically "rules of thumb" that help us figure out what to do next when we don't have the perfect answer. Heuristics can come from many different places: intuition, experience, and other people's opinions are good sources for heuristic solutions. The only requirement is that they be effective enough to solve the problem at hand!