Machine Learning V/S Deep Learning

 

Artificial intelligence has two main subfields: machine learning and deep learning, each having special traits of its own. Deep learning, a subtype of machine learning, uses sophisticated neural networks that are modeled after the human brain to process massive datasets, whereas machine learning utilizes algorithms that identify patterns from data without the need for explicit programming. The table below illustrates the key disparities between the two subfields of AI. 

Definition	a branch of artificial intelligence where algorithms use data to learn and make decisions or predictions	Artificial neural networks are a subset of machine learning that use vast quantities of data (labeled and unlabeled data) to learn.
Complexity	usually simpler models with less parameters.	involves intricate structures with a large number of layers and factors.
Feature Engineering	manual feature extraction and selection is required.	Automatically learns features from raw data
Representation	makes use of simple models, such as support vector machines and decision trees.	Relies heavily on artificial neural networks
Performance	May struggle with complex tasks and large and complex datasets	Good at handling complex tasks and large datasets
Training Data	Requires labeled data for supervised learning, reinforcement learning	relies little on labeled data. It can benefit from large amounts of unlabeled data too.
Interpretability	Simpler and interpretable models	less interpretable