Models Vs. Algorithms
There is no way of explaining how AI works without mentioning “model ”“and “algorithm, these terms are what make the process work. From a far it may seem like they are one in the same because of how intertwined they are, but in reality they’re quite different. An algorithm is like a recipe that contains ingredients and directions on how to make the dish, while the model is the final product (completed dish).
Let’s look at it a bit closer, starting with Algorithms:
Algorithms are precise sets of instructions that guide computers in performing specific tasks. These instructions can range from simple calculations to complex decision-making processes. In machine learning, algorithms are employed to analyze vast amounts of data, discover underlying patterns, and construct models capable of making informed predictions or decisions.
Humans use algorithms in theri daily life, the process doesn’t just related to computer science. To put it into perspective, Algorithms are step-by-step procedures designed to solve a specific problem or achieve a particular goal.
Real life examples include:
Sorting Papers:
Although it is a simple task of diving paper into different categories, the movements follow a set of steps to achieve an outcome.
Sorting papers shows the variety of tasks and specifications algorithms can use. For instance, you can sort your files alphabetically, by word count, date, and countless others. The goal is to simplify the organizational process by using small tasks.
Traffic Signals
Believe it or not algorithms can be detected in traffic signals. Every movement is grouped into phrases.
For example, driving through a lane and turning right would be grouped into one phase.
To maintain safety, traffic lights need a smart algorithm that detects the phases and times movements are completed correctly.
Every time a person stops at a red light, the traffic signal cycles through phases. An algorithm can also be used to determine the best time a person can continue driving through an intersection based on traffic volume.
Models:
A model learns from what was generated in a machine learning by the algorithm. Different algorithms are applied to relevant data inputs in order to achieve the tasks they’ve been programmed for. In a few words, a model is used to make predictions or decisions.
A human like example of how models work can be compared with students in a history class. Let’s say students are learning about the civil war:
Students ask their probing questions about the topic, the professor then analyzes the questions, identifies key concepts, and formulates relevant questions / information they were asked in the past to draw upon and compare with their immense knowledge base of the civil war..
They proceed to challenge the student’s assumptions / inquiries and encourage critical thinking.
In the end, the professor can provide a variety of information related to the student’s question or create a new perspective.
Hope this helps!