Machine Learning Tutorial 1- Difference between Machine Learning and Traditional Programming|dileenet A/L ICT

Difference between Machine Learning and Traditional Programming

 

Machine learning and traditional programming are two distinct approaches to solving problems and building applications.

Here are the key differences between them:

 

	Traditional Programming	Machine Learning
Approach to Problem Solving:	developers explicitly write a set of rules and instructions (algorithms) to solve a specific problem or achieve a particular task.	the algorithm learns patterns and rules from data to make predictions or decisions without being explicitly programmed.
Data Dependency	data-independent; the code's behavior is determined solely by the written instructions, not by any specific data.	data-dependent; the model's behavior and performance heavily rely on the training data it's exposed to.
Rules Generations	creating specific rules and conditions based on human understanding and expertise.	learn rules and patterns from data through training and adjust their behavior accordingly.
Flexibility and Adaptability:	generally static and require manual modification of the code to accommodate changes or adapt to new data patterns.	Machine learning models can adapt and improve automatically based on new data, making them more flexible in handling changing scenarios.
Performance Improvement:	performance improvements often require manual optimization of algorithms or code.	performance improvement can be achieved by retraining the model with additional data or fine-tuning the existing model parameters.
Programming Paradigm:	follows a deterministic paradigm, where the output is entirely determined by the input and the program's logic.	follows a statistical or probabilistic paradigm, where the output is based on learned patterns and statistical probabilities.
Human Involvement:	heavily relies on human expertise and domain knowledge to design the algorithms and rules.	involves less explicit human intervention in designing the specific rules, as the model learns from data.
Error Handling:	errors and exceptions are explicitly handled through code, based on predefined rules and conditions.	errors are learned from the training data, and the model adjusts its predictions based on these errors during training.
Use Cases:	suitable for rule-based systems, where the problem and solution can be explicitly defined and understood.	highly effective for complex tasks like image and speech recognition, natural language processing, recommendation systems, and other tasks where patterns need to be learned from data.

 

How machine learning algorithms work

Forward Pass:

 In the Forward Pass, the machine learning algorithm takes in input data and produces an output. Depending on the model algorithm it computes the predictions.

Loss Function:

The loss function, also known as the error or cost function, is used to evaluate the accuracy of the predictions made by the model. The function compares the predicted output of the model to the actual output and calculates the difference between them. This difference is known as error or loss. The goal of the model is to minimize the error or loss function by adjusting its internal parameters.

Model Optimization Process:

The model optimization process is the iterative process of adjusting the internal parameters of the model to minimize the error or loss function. This is done using an optimization algorithm, such as gradient descent. The optimization algorithm calculates the gradient of the error function with respect to the model’s parameters and uses this information to adjust the parameters to reduce the error. The algorithm repeats this process until the error is minimized to a satisfactory level.

How Does Machine Learning Work in Advertising

Advertising platforms want to match each potential customer with the best possible ad for their situation, as personalization can help drive better advertising performance. Facebook uses machine learning heavily in its recommendation algorithm: Its model incorporates existing behavioral data from each user and predicts the types of advertisements most beneficial for each individual. This is a great use case for machine learning because of the large amount of data that is processed. The recommendations also continue to improve over time, as more data is fed into the system, allowing for more personalized advertising for every user.