Backpropagation

TensorFlow

Feature/Independent variable

Image and video processing

Pooling layer

Bag of Words

Sequential data

Tokenization

Automatically learn spatial hierarchies

Map words into numerical vectors

Computer Vision, Speech Recognition, Natural Language Processing

Convolutional Layer, Pooling Layer, Flatten Layer

One-to-one, One-to-many, Many-to-one, Many-to-many

Data requirements, Need for Feature Engineering, Interpretability, Model complexity

Tokenization, Padding, Vectorization