Big Data II Machine Learning and Deep Learning

deep learning definition

Deep learning refers to a type of machine learning that allows computers to learn complex concepts by learning simpler concepts and combining them

K-means

an algorithm in which "k" indicates the number of clusters and "means" represents the clusters' centroids

recurrent neural networks

any network with neurons that send feedback signals to each other s(t)= f(s(t-1), x (t);0theta -one to one, one to many, many to one, many to many, many to many

Support Vector Machine

can extrapolate information from one dimensional data (input space) and some information about weights & correlative relationships to another dimension (feature space) -maximize with of the decision boundary

neural networks

input layer, hidden layer, output layer

deep learning

the most advance machine learning techniques

Convolutional neural networks

use kernels that look at small portions of the image separately first •A kernel is a small matrix used for convolution in image processing -input image, convolutional layer, pooling layer, convolution layer, pooling, conv, pooling, connected layers, software output

linear model =

y= b0+b1x

Together RNNs and CNNs can be used for:

•Image captioning •Video classification •Video generation

RNNs can be used for:

•Modelling the stock market •Natural Language Processing (NLP) •Supply chain forecasting •Machine translation •Speech processing

deep neural network

•Refers to a neural network with more than one hidden layer •Two most common DNNs are: •Convolutional neural network (CNN) - used for image classification •Recurrent neural networks (RNN) - used for natural language processing and for sequential data

deep learning uses layers

•Some layers capture simple concepts •Other layers build on the simpler concepts -input layer, hidden layers output layer

three types of machine learning

•Supervised learning •Unsupervised learning •Reinforcement learning

Deep neural networks are trained through a process called optimization

•This is like regression, where we try to minimize the loss function •The parameters used to tune neural network optimization are called hyperparameters

Convolutional neural networks can be used for:

Image classification •Video classification •Image captioning -Natural language processing

machine learning

In 1959 Arthur Samuel coined the term machine learning •He defined machine learning as a:•Field of study that gives computers the ability to learn without being explicitly programmed. •In 1998 Tom Mitchell proposed a more precise definition: •A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.

logistic model=

P= 1/ ( 1+e)^-(b0+b1x)

Advanced RNNs include:

Recursive neural networks (RNN) •Gated recurrent units (GRUs) •Long short-term memories (LSTMs)

reinforcement

-decision process -reward system -learn series of action

supervised

-labeled data -direct feedback -predict future outcome -goes to classification and regression on a squiggly graph with a boundary

unsupervised

-no labels -no feedback -fine hidden structure -goes to clustering