| # | Chapter | Link |
|---|---|---|
| 1 | Biological Neuron | Download
Verified |
| 2 | From Spring to Winter of AI | Download
Verified |
| 3 | The Deep Revival | Download
Verified |
| 4 | From Cats to Convolutional Neural Networks | Download
Verified |
| 5 | Faster, higher, stronger | Download
Verified |
| 6 | The Curious Case of Sequences | Download
Verified |
| 7 | Beating humans at their own games (literally) | Download
Verified |
| 8 | The Madness (2013-) | Download
Verified |
| 9 | (Need for) Sanity | Download
Verified |
| 10 | Motivation from Biological Neurons | Download
Verified |
| 11 | McCulloch Pitts Neuron, Thresholding Logic | Download
Verified |
| 12 | Perceptrons | Download
Verified |
| 13 | Error and Error Surfaces | Download
Verified |
| 14 | Perceptron Learning Algorithm | Download
Verified |
| 15 | Proof of Convergence of Perceptron Learning Algorithm | Download
Verified |
| 16 | Deep Learning(CS7015): Linearly Separable Boolean Functions | Download
Verified |
| 17 | Deep Learning(CS7015): Representation Power of a Network of Perceptrons | Download
Verified |
| 18 | Deep Learning(CS7015): Sigmoid Neuron | Download
Verified |
| 19 | Deep Learning(CS7015): A typical Supervised Machine Learning Setup | Download
Verified |
| 20 | Deep Learning(CS7015): Learning Parameters: (Infeasible) guess work | Download
Verified |
| 21 | Deep Learning(CS7015): Learning Parameters: Gradient Descent | Download
Verified |
| 22 | Deep Learning(CS7015): Representation Power of Multilayer Network of Sigmoid Neurons | Download
Verified |
| 23 | Feedforward Neural Networks (a.k.a multilayered network of neurons) | Download
Verified |
| 24 | Learning Paramters of Feedforward Neural Networks (Intuition) | Download
Verified |
| 25 | Output functions and Loss functions | Download
Verified |
| 26 | Backpropagation (Intuition) | Download
Verified |
| 27 | Backpropagation: Computing Gradients w.r.t. the Output Units | Download
Verified |
| 28 | Backpropagation: Computing Gradients w.r.t. Hidden Units | Download
Verified |
| 29 | Backpropagation: Computing Gradients w.r.t. Parameters | Download
Verified |
| 30 | Backpropagation: Pseudo code | Download
Verified |
| 31 | Derivative of the activation function | Download
Verified |
| 32 | Information content, Entropy & cross entropy | Download
Verified |
| 33 | Recap: Learning Parameters: Guess Work, Gradient Descent | Download
Verified |
| 34 | Contours Maps | Download
Verified |
| 35 | Momentum based Gradient Descent | Download
Verified |
| 36 | Nesterov Accelerated Gradient Descent | Download
Verified |
| 37 | Stochastic And Mini-Batch Gradient Descent | Download
Verified |
| 38 | Tips for Adjusting Learning Rate and Momentum | Download
Verified |
| 39 | Line Search | Download
Verified |
| 40 | Gradient Descent with Adaptive Learning Rate | Download
Verified |
| 41 | Bias Correction in Adam | Download
Verified |
| 42 | Eigenvalues and Eigenvectors | Download
Verified |
| 43 | Linear Algebra : Basic Definitions | Download
Verified |
| 44 | Eigenvalue Decompositon | Download
Verified |
| 45 | Principal Component Analysis and its Interpretations | Download
Verified |
| 46 | PCA : Interpretation 2 | Download
Verified |
| 47 | PCA : Interpretation 3 | Download
Verified |
| 48 | PCA : Interpretation 3 (Contd.) | Download
Verified |
| 49 | PCA : Practical Example | Download
Verified |
| 50 | Singular Value Decomposition | Download
Verified |
| 51 | Introduction to Autoncoders | Download
Verified |
| 52 | Link between PCA and Autoencoders | Download
Verified |
| 53 | Regularization in autoencoders (Motivation) | Download
Verified |
| 54 | Denoising Autoencoders | Download
Verified |
| 55 | Sparse Autoencoders | Download
Verified |
| 56 | Contractive Autoencoders | Download
Verified |
| 57 | Bias and Variance | Download
Verified |
| 58 | Train error vs Test error | Download
Verified |
| 59 | Train error vs Test error (Recap) | Download
Verified |
| 60 | True error and Model complexity | Download
Verified |
| 61 | L2 regularization | Download
Verified |
| 62 | Dataset augmentation | Download
Verified |
| 63 | Parameter sharing and tying | Download
Verified |
| 64 | Adding Noise to the inputs | Download
Verified |
| 65 | Adding Noise to the outputs | Download
Verified |
| 66 | Early stopping | Download
Verified |
| 67 | Ensemble Methods | Download
Verified |
| 68 | Dropout | Download
Verified |
| 69 | A quick recap of training deep neural networks | Download
Verified |
| 70 | Unsupervised pre-training | Download
Verified |
| 71 | Better activation functions | Download
Verified |
| 72 | Better initialization strategies | Download
Verified |
| 73 | Batch Normalization | Download
Verified |
| 74 | One-hot representations of words | Download
Verified |
| 75 | Distributed Representations of words | Download
Verified |
| 76 | SVD for learning word representations | Download
Verified |
| 77 | SVD for learning word representations (Contd.) | Download
Verified |
| 78 | Continuous bag of words model | Download
Verified |
| 79 | Skip-gram model | Download
Verified |
| 80 | Skip-gram model (Contd.) | Download
Verified |
| 81 | Contrastive estimation | Download
Verified |
| 82 | Hierarchical softmax | Download
Verified |
| 83 | GloVe representations | Download
Verified |
| 84 | Evaluating word representations | Download
Verified |
| 85 | Relation between SVD and Word2Vec | Download
Verified |
| 86 | The convolution operation | Download
Verified |
| 87 | Relation between input size, output size and filter size | Download
Verified |
| 88 | Convolutional Neural Networks | Download
Verified |
| 89 | Convolutional Neural Networks (Contd.) | Download
Verified |
| 90 | CNNs (success stories on ImageNet) | Download
Verified |
| 91 | CNNs (success stories on ImageNet) (Contd.) | Download
Verified |
| 92 | Image Classification continued (GoogLeNet and ResNet) | Download
Verified |
| 93 | Visualizing patches which maximally activate a neuron | Download
Verified |
| 94 | Visualizing filters of a CNN | Download
Verified |
| 95 | Occlusion experiments | Download
Verified |
| 96 | Finding influence of input pixels using backpropagation | Download
Verified |
| 97 | Guided Backpropagation | Download
Verified |
| 98 | Optimization over images | Download
Verified |
| 99 | Create images from embeddings | Download
Verified |
| 100 | Deep Dream | Download
Verified |
| 101 | Deep Art | Download Verified |
| 102 | Fooling Deep Convolutional Neural Networks | Download Verified |
| 103 | Sequence Learning Problems | Download Verified |
| 104 | Recurrent Neural Networks | Download Verified |
| 105 | Backpropagation through time | Download Verified |
| 106 | The problem of Exploding and Vanishing Gradients | Download Verified |
| 107 | Some Gory Details | Download Verified |
| 108 | Selective Read, Selective Write, Selective Forget - The Whiteboard Analogy | Download Verified |
| 109 | Long Short Term Memory(LSTM) and Gated Recurrent Units(GRUs) | Download Verified |
| 110 | How LSTMs avoid the problem of vanishing gradients | Download Verified |
| 111 | How LSTMs avoid the problem of vanishing gradients (Contd.) | Download Verified |
| 112 | Introduction to Encoder Decoder Models | Download Verified |
| 113 | Applications of Encoder Decoder models | Download Verified |
| 114 | Attention Mechanism | Download Verified |
| 115 | Attention Mechanism (Contd.) | Download Verified |
| 116 | Attention over images | Download Verified |
| 117 | Hierarchical Attention | Download Verified |