Personal tools

Foundations of DL

Machine Learning Vs Deep Learning_122723A
[Machine Learning Vs Deep Learning - Semiconductor Engineering]

- Overview

Deep learning is a type of machine learning (ML) research that uses artificial neural networks (ANNs) to conduct automated data analysis. It's a subset of machine learning (ML) that's based on representation learning and artificial neural networks (ANNs).

DL is used in many fields, including: 

  • Computer vision
  • Speech recognition
  • Natural language processing
  • Machine translation
  • Bioinformatics
  • Drug design
  • Medical image analysis
  • Climate science
  • Material inspection
  • Board game programs

DL techniques can capture complex relations between non-related fields. For example, they can capture the relationship between air pressure recordings and English words, or between millions of pixels and a textual description. 

DL is a combination of neural networks, AI, graphical modeling, optimization, pattern recognition, and signal processing.

ML is about computers being able to think and act with less human intervention. DL is about computers learning to think using structures modeled on the human brain. 


- DL Algorithms

Deep learning (DL) is a subset of ML. It's a type of AI that uses neural networks to learn from large amounts of data. 

Neural networks are made up of interconnected nodes, or neurons, that are layered to resemble the human brain. They mimic how neurons in the brain signal each other, which is why they're called "neural". 

DL models can be taught to perform classification tasks and recognize patterns in text, photos, audio, and other data. 

Here are some DL algorithms: 

  • Convolutional neural network: Uses filters to learn the features of an image, such as important objects.
  • Recurrent neural network: Uses a sequential approach and performs mathematical calculations in a sequence.
  • Generative adversarial network (GAN): A class of algorithms that consists of two adversarial networks. One network generates realizations, and the other tries to differentiate real from simulated data.
  • Autoencoder: A three-layer neural network that tries to reconstruct the input with minimal error.
  • Multilayer perceptron (MLP): A deep learning method that helps with complex computations and increases the prediction accuracy of the training model.
  • Decision tree: Uses machine and deep learning to automate complex business processes.
  • k-Nearest Neighbor (kNN) classification algorithm: A simple classification algorithm that uses deep learning to classify by measuring the distance between different feature values.
  • Logistic regression: Used as a classifier in the final layer of a deep learning. It is fast and simple, so it is used for large datasets.
  • Cluster analysis: A clustering algorithm that divides data based on similarities. The grouped data are similar to each other more than the other data in other groups.



[More to come ...]


Document Actions