Traditional Algorithms are designed for problems which we have already know how to solve and so we explicitly program a computer to solve them.
While machine learning Algorithm deals with a problem without clear-cut steps towards a solution.
The output of a machine learning Algorithm varies with the hyperparameters such as training data, testing data, training time, learning rate, optimizer, epoch etc.
Machine Learning in layman definition: is the ability of a system to learn and improve from experience (data) without being explicitly programmed. or is the way to make computers learn how to perform complex tasks whose processes cannot be easily described by humans.
Machine learning Algorithm is the ability of a computer program to search for patterns in data to make better decisions or a set of procedures that help a model adapt to the data given or specifies the way the data is transformed from input to output and how the model learns the appropriate mapping from input to output.
Machine Learning Techniques/Style learning style is the ways an algorithm model a problem based on its interaction with the experience or environment. Learning Style an Algorithm can Adopt is Supervised Learning, Unsupervised learning, Semi-supervised and Reinforcement Learning.
List of Most used Algorithms
Support Vector Machine
CART (Classification And Regression Tree)
Principal Component Analysis (PCA) for dimensionality reduction Algorithm.
Iterative Dichotomiser 3 (ID3)
Gradient Boosting (XGBoost and LightGBM)
Real-world applications of Machine learning
Stock Trading prediction (Recurrent Neural Network)
Web search and recommendation engine e.g Google search and google map, Netflix and Amazon product recommendation
Drug discovery and Computational Biology (Generative Adversarial Networks)
Credit card Fraud detection e.g Paypal credit card fraud detection.
Face detection and Recognition e.g Facebook Face recognition and friend suggestions
Spam detection e.g Gmail spam classification
A medical diagnosis for detecting diseases
Suspicious activity detection from a CCTV camera. ( computer vision)
Speech Understanding and conversion from one language to another e.g Apple Siri, Microsoft Cortana, Amazon Echo.
Note: Deep learning is just a machine learning Algorithm as Logistic Regression is.