we implement a transformer architecture to predict whether tweets are worthy or subpar

we set up a net with convolutional architecture in pytorch, which can distinguish images of birds from those of airplanes

we construct a neural net for sentiment prediction from scratch using numpy

we use a convolutional network to predict the inputs and outputs of a separate target net

we show how to use a convolutional net for the identify-handwritten-numbers task

we derive the backprop equations necessary for training a convolutional neural network

we derive the backpropagation equations used to correct the weights for a 3 layer neural net with an arbitrary number of neurons

We create a three layer neural net that can identify handwritten numbers and use a regularization technique called dropout to prevent overfitting

we can best understand how neural nets work by studying a small, simplified example

unsupervised learning algorithms such as Isomap and K-means can produce intuitive categorizations for unlabeled data