Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Foundation of Deep Learning

sambaiga
April 04, 2018

Foundation of Deep Learning

The tutorial present the foundations principles of deep learning from both theoretical and implementation perspective. Specifically it presents the basic building block of deep learning model, modern deep learning architectures and their application and how to train deep learning algorithms with emphasis on techniques used in practise.

sambaiga

April 04, 2018
Tweet

More Decks by sambaiga

Other Decks in Technology

Transcript

  1. Foundation of Deep Learning PytzMLS2018@IdabaX: CIVE UDOM Tanzania. Anthony Faustine

    PhD Fellow (IDLab research group-Ghent University) 4 April 2018 1
  2. Learning goal • Understand the basic building block of deep

    learning model. • Learn how to train deep learning models. • Learn different techniques used in practise to train deep learning models. • Understand different modern deep learning architectures and their application. • Explore opportunities and research direction in deep learning. 2
  3. Outline Introduction to Deep learning Multilayer Perceptron Training Deep neural

    networks Deep learning in Practice Modern Deep learning Architecture Limitation and Research direction of deep neural networks 3
  4. What is Deep Learning Deep Learning a subclass of machine

    learning algorithms that learn underlying features in data using multiple processing layers with multiple levels of abstarction. Figure 1: ML vs Deep learning: credit: 3
  5. Deep Learning Success Automatic Colorization Figure 2: Automatic colorization Object

    Classification and Detection Figure 3: Object recognition Image Captioning Image Style Transfer 4
  6. Deep Learning Success Pneumonia Detection on Chest X-Rays Pedict heart

    disease risk from eye scans Computational biology Diagnosis of Skin Cancer More stories 7
  7. Why Deep Learning and why now? Why deep learning: Hand-Engineered

    Features vs. Learned features. Traditional ML • Use enginered feature to extract useful patterns from data. • Complex and difficult since different data sets require different feature engineering approach Deep learning • Automatically discover and extract useful pattern from data. • Allows learning complex features e.g speach and complex networks. 8
  8. Why Deep Learning and why now? Why Now? Big data

    availability • Large datasets • Easier collection and storage Increase in computaional power • Modern GPU architecture. Improved techniques • Five decades of research in machine learning. Open source tools and models • Tensorflow. • Pytorch • Keras 9
  9. Outline Introduction to Deep learning Multilayer Perceptron Training Deep neural

    networks Deep learning in Practice Modern Deep learning Architecture Limitation and Research direction of deep neural networks 10
  10. The Perceptron A perceptron is a simple model of a

    neuron. The output: ˆ y = f(x) = g(z(x)) where • x, y input, output. • w, b weight and bias parameter θ • activation function: g(.) • pre-activation: z(x) = n i=1 wixi + b 10
  11. Perceptron ˆ y = g(z(x)) ˆ y = g(b +

    n i=1 wixi)) ˆ y = g(b + wx)) 11
  12. The Perceptron: Activation Function Why Activation Functions? • Activation functions

    add non-linearity properties to neuro network function. • Most real-world problems + data are non-linear. • Activation function need to be differentiable. Figure 4: Activation function credit:kdnuggets.com 12
  13. Multilayer Perceptrons (MLP) • Consists of L multiple layers (l1

    , l2 . . . lL ) of pecepron, interconnected in a feed-forward way. • The first layer l1 is called the input layer ⇒ just pass the information to the next layer. • The last layer is the ouput layer ⇒ maps to the desired output format. • The intermediate k layers are hidden layers ⇒ perform computations and transfer the weights from the input layer. 14
  14. Multilayer Perceptrons (MLP) • Input: x = {x1, x2, .

    . . xd} ∈ R(d×N) • Pre-activation: z(1)(x) = b(1) + w(1)(x) where z(x)i = j w(1) i,j xj + b(1) i Hidden layer 1 • Activation h(1)(x) = g(z(1)(x)) = g(b(1) + w(1)(x)) • Pre-activation z(2)(x) = b(2) + w(2)h(1)(x) 15
  15. Multilayer Perceptrons (MLP) Hidden layer 2 • Activation h(2)(x) =

    g(z(2)(x)) = g(b(2) + w(2)h(1)(x)) • Pre-activation z(3)(x) = b(3) + w(3)h(2)(x) Hidden layer k • Activation h(k)(x) = g(z(k)(x)) = g(b(k) + w(k)h(k−1)(x)) • Pre-activation z(k+1)(x) = b(k+1) + w(k+1)h(k)(x) 16
  16. Multilayer Perceptrons (MLP) Output layer • Activation h(k+1)(x) = O(z(k+1)(x))

    = O(b(k+1) + w(k+1)h(k)(x)) = ˆ y where O(.) is output activation function Output activation function • Binary classification: y ∈ {0, 1} ⇒ sigmoid • Multiclass classfiction:y ∈ {0, K − 1} ⇒ softmax • Regression: y ∈ Rn ⇒ identity sometime RELU. Demo Playground 17
  17. MLP: Pytorch import torch model = torch.nn.Sequential( torch.nn.Linear(2, 16), torch.nn.ReLU(),

    torch.nn.Linear(16, 64), torch.nn.ReLU(), torch.nn.Linear(64, 1024), torch.nn.ReLU(), torch.nn.Linear(1024, 1), torch.nn.Sigmoid() ) 18
  18. MLP: Pytorch import torch from torch.nn import functional as F

    class MLP(torch.nn.Module): def __init__(self): super(MLP, self).__init__() self.fc1 = torch.nn.Linear(2, 16) self.fc2 = torch.nn.Linear(16, 64) self.fc3 = torch.nn.Linear(64, 1024) self.out = torch.nn.Linear(1024, 1) def forward(self, x): x = F.relu(self.fc1(x)) x = F.relu(self.fc2(x)) x = F.relu(self.fc3(x)) out = F.sigmoid(self.out(x)) return x model = MLP() 19
  19. Outline Introduction to Deep learning Multilayer Perceptron Training Deep neural

    networks Deep learning in Practice Modern Deep learning Architecture Limitation and Research direction of deep neural networks 20
  20. Training Deep neural networks To train DNN we need: 1

    Define loss function: L(f(x(i) : θ), y(i)) 2 A procedure to compute gradient ∂Jθ ∂θ 3 Solve optimisation problem. 20
  21. Training Deep neural networks: Define loss function The type of

    Loss function is determined by the output layer of MLP. Binary classification Output • Predict y ∈ {0, 1} • Use sigmoid σ(.) activation function. p(y = 1|x) = 1 1 + e−x Loss • Binary cross entropy. L(ˆ y, y) = y log ˆ y − (1 − y) log(1 − ˆ y) • torch.nn.BCELoss() 21
  22. Training Deep neural networks: Define loss function Mutli class classification

    Output • Predict y ∈ {1, k} • Use softmax σ(.) activation function. p(y = i|x) = exp(xi ) k j Loss • Cross entropy. L(ˆ y, y) = k i=1 yi log ˆ yi • torch.nn.CrossEntropyLoss() 22
  23. Training Deep neural networks: Define loss function Regression Output •

    Predict y ∈ Rn • Use identity activation function and sometime ReLU activation. Loss • Squared error loss. L(ˆ y, y) = 1 2 (yi − ˆ yi )2 • torch.nn.MSELoss() 23
  24. Training Deep neural networks: Compute Gradients Backpropagation: a procedure that

    is used to compute gradients of a loss function. • It is based on the application of the chain rule and computationally proceeds ’backwards’. Figure 5: Back propagation: credit: Flair of Machine Learnin 24
  25. Training Deep neural networks: Backpropagation Consider a following single hidden

    layer MLP. Forward path z = w1x + b1 h = g(z) ˆ y = w2h + b2 Jθ = 1 2 ||y − ˆ y||2 We need to find: ∂Jθ ∂w(1) , ∂Jθ ∂b(1) , ∂Jθ ∂w(2) and ∂Jθ ∂b(2) 25
  26. Training Deep neural networks: Backpropagation Back ward path Jθ =

    1 2 ||y − ˆ y||2 ∂Jθ ∂ˆ y = ||y − ˆ y|| 26
  27. Training Deep neural networks: Backpropagation Back ward path ˆ y

    = w2h + b2 ∂Jθ ∂w(2) = ∂ˆ y ∂w(2) · ∂Jθ ∂ˆ y = hT · ||y − ˆ y|| ∂Jθ ∂b(2) = ∂ˆ y ∂b(2) · ∂Jθ ∂ˆ y = ||y − ˆ y|| 27
  28. Training Deep neural networks: Backpropagation Back ward path ˆ y

    = w2h + b2 h = g(z) ∂Jθ ∂h = ∂ˆ y ∂h · ∂Jθ ∂ˆ y = w(2)T · ||y − ˆ y|| ∂Jθ ∂z = ∂h ∂z · ∂Jθ ∂h = g ((z)) · ∂Jθ ∂h 28
  29. Training Deep neural networks: Backpropagation Back ward path z =

    w1h + b1 ∂Jθ ∂w(1) = ∂z ∂w(1) · ∂Jθ ∂z = xT · ∂Jθ ∂z ∂Jθ ∂b(1) = ∂z ∂b(1) · ∂Jθ ∂z = ∂Jθ ∂z 29
  30. Training Neural Networks: Solving optimisation problem Objective: Find parameters θ

    : w and b that minimize the cost function: arg max θ 1 N i L(f(x(i) : θ), y(i)) Figure 6: Visualizing the loss landscape of neural nets: credit: Hao Li 30
  31. Training Neural Networks: Gradient Descent Gradient Descent 1 Initilize parameter

    θ, 2 Loop until converge 1 Compute gradient: ∂Jθ ∂θ 2 Update parameters: θt+1 = θt − α ∂Jθ ∂θ 3 Retrn parameter θ Limitation: Take time to compute 31
  32. Training Neural Networks: Stochastic Gradient Descent (SGD) SGD consists of

    updating the model parameters θ after every sample. SGD Initialize θ randomly. For each training example: • Compute gradients: ∂Jiθ ∂θ • Update parameters θ with update rule: θ(t+1) := θ(t) − α ∂Jiθ ∂θ Stop when reaching criterion Easy to compute ∂Jiθ ∂θ but very noise. 32
  33. Training Neural Networks: Mini-batch SGD training Make update based on

    a min-batch B of example instead of single example i Mini-batch SGD 1 Initialize θ randomly. 2 For each mini-batch B: • Compute gradients: ∂Jθ ∂θ = 1 B B k=1 ∂Jk(θ) ∂θ • Update parameters θ with update rule: θ(t+1) := θ(t) − α∂Jiθ ∂θ 3 Stop when reaching criterion Fast to compute ∂Jθ ∂θ = 1 B B k=1 ∂Jk(θ) ∂θ and much better estimate of the true gradient. Standard procedure for training deep learning. 33
  34. Training Neural Networks: Gradient Descent Issues Setting the learning rate

    α • Small learning rate: Converges slowly and gets stuck in false local minima. • Large learning rate: Overshoot became unstable and diverge. • Stable learning rate: Converges smoothly and avoid local minima. How to deal with this ? 1 Try lots of different learning rates and see what works for you. • Jeremy propose a technique to find stable learning rate 2 Use an adaptive learning rate that adapts to the landscape of your loss function. 34
  35. Training Neural Networks: Adaptive Learning rates algorithm 1 Momentum 2

    Adagrad 3 Adam 4 RMSProp pytorch optimer algorithms 35
  36. Outline Introduction to Deep learning Multilayer Perceptron Training Deep neural

    networks Deep learning in Practice Modern Deep learning Architecture Limitation and Research direction of deep neural networks 36
  37. Deep learning in Practice: Regularization Regularization: Technique to help deep

    learning network perform better on unsee data. • Constraints optimization problem to discourage complex model. arg max θ 1 N i L(f(x(i) : θ), y(i)) + λΩ(θ) • Improve generalization of deep learning model. 36
  38. Regularization 1: Dropout Dropout: Randomly remove hidden unit from a

    layer during training step and put them back during test. • Each hidden unit is set to 0 with probability p. • Force network to not rely on any hidden node ⇒ prevent neural net from ovefitting (improve performance). • Any dropout probability can be used but 0.5 usually works well. 37
  39. Regularization 1: Dropout Dropout: in pytorch is implemented as torch.nn.Dropout

    If we have a network: model = torch.nn.Sequential( torch.nn.Linear(1,100), torch.nn.ReLU(), torch.nn.Linear(100,50), torch.nn.ReLU(), torch.nn.Linear(50,2)) We can simply add dropout layers: model = torch.nn.Sequential( torch.nn.Linear(1,100), torch.nn.ReLU(), torch.nn.Dropout() torch.nn.Linear(100,50), torch.nn.ReLU(), torch.nn.Dropout() torch.nn.Linear(50,2)) Note: A model using dropout has to be set in train or eval model. 38
  40. Regularization 1: Dropout Dropout: in pytorch is implemented as torch.nn.Dropout

    If we have a network: model = torch.nn.Sequential( torch.nn.Linear(1,100), torch.nn.ReLU(), torch.nn.Linear(100,50), torch.nn.ReLU(), torch.nn.Linear(50,2)) We can simply add dropout layers: model = torch.nn.Sequential( torch.nn.Linear(1,100), torch.nn.ReLU(), torch.nn.Dropout() torch.nn.Linear(100,50), torch.nn.ReLU(), torch.nn.Dropout() torch.nn.Linear(50,2)) Note: A model using dropout has to be set in train or eval model. 38
  41. Regularization 2: Early Stopping Early Stopping: Stop training before the

    model overfit. • Monitor the deep learning training process from overfiting. • Stop training when validation error increases. Figure 7: Early stopping: credit: Deeplearning4j.com 39
  42. Deep learning in Practice: Batch Normalization Batch normalisation: A technique

    for improving the performance and stability of deep neural networks. Training deep neural network is complicated • The input of each layer changes as the parameter of the previous layer change. • This slow down the training ⇒ require low learning rate and careful parameter initilization. • Make hard to train models with saturation non-linearity. • This phenomena is called Covariate shift To address covariate shift ⇒ normalise the inputs of each layer for each mini-batch (Batch normalization) • To have a mean output activation of zero and standard deviation of one. 40
  43. Deep learning in Practice: Batch Normalization Batch normalisation: A technique

    for improving the performance and stability of deep neural networks. Training deep neural network is complicated • The input of each layer changes as the parameter of the previous layer change. • This slow down the training ⇒ require low learning rate and careful parameter initilization. • Make hard to train models with saturation non-linearity. • This phenomena is called Covariate shift To address covariate shift ⇒ normalise the inputs of each layer for each mini-batch (Batch normalization) • To have a mean output activation of zero and standard deviation of one. 40
  44. Deep learning in Practice: Batch Normalization If x1, x2, .

    . . xB are the sample in the batch with mean ˆ µb and variance ˆ σ2 b . • During training batch normalization shift and rescale each component of the input according to batch statistics to produce output yb : yb = γ xb − ˆ µb ˆ σ2 b + + β where • is the Hadamard component-wise product. • The parameter γ and β are the desired moments which are either fixed or optimized during training. • As for dropout the model behave differently during training and test. 41
  45. Deep learning in Practice: Batch Normalization Batch Normalization: in pytorch

    is implemented as torch.nn.BatchNorm1d If we have a network: model = torch.nn.Sequential( torch.nn.Linear(1,100), torch.nn.ReLU(), torch.nn.Linear(100,50), torch.nn.ReLU(), torch.nn.Linear(50,2)) We can simply add batch normalization layers: model = torch.nn.Sequential( torch.nn.Linear(1,100), torch.nn.ReLU(), torch.nn.BatchNorm1d(100) torch.nn.Linear(100,50), torch.nn.ReLU(), torch.nn.BatchNorm1d(50) torch.nn.Linear(50,2)) Note: A model using batch has to be set in train or eval model. 42
  46. Deep learning in Practice: Batch Normalization Batch Normalization: in pytorch

    is implemented as torch.nn.BatchNorm1d If we have a network: model = torch.nn.Sequential( torch.nn.Linear(1,100), torch.nn.ReLU(), torch.nn.Linear(100,50), torch.nn.ReLU(), torch.nn.Linear(50,2)) We can simply add batch normalization layers: model = torch.nn.Sequential( torch.nn.Linear(1,100), torch.nn.ReLU(), torch.nn.BatchNorm1d(100) torch.nn.Linear(100,50), torch.nn.ReLU(), torch.nn.BatchNorm1d(50) torch.nn.Linear(50,2)) Note: A model using batch has to be set in train or eval model. 42
  47. Deep learning in Practice: Batch Normalization When Applying Batch Normalization

    • Carefully shuffle your sample. • Learning rate can be greater. • Dropout is not necessary. • L2 regularization influence should be reduced. 43
  48. Deep learning in Practice: Weight Initilization Before training the neural

    network you have to initialize its parameters. Set all the initial weights to zero • Every neuron in the network will computes the same output ⇒ same gradients. • Not recommended 44
  49. Deep learning in Practice: Weight Initilization Random Initilization • Initilize

    your network to behave like zero-mean standard gausian function. wi ∼ N µ = 0, σ = 1 n bi = 0 where n is the number of inputs. 45
  50. Deep learning in Practice: Weight Initilization Random Initilization: Xavier initilization

    • Initilize your network to behave like zero-mean standard gausian function such that wi ∼ N µ = 0, σ = 1 nin + nout bi = 0 where nin, nout are the number of units in the previous layer and the next layer respectively. where n is the number of inputs. 46
  51. Deep learning in Practice: Weight Initilization Random Initilization: Kaiming •

    Random initilization that take into account ReLU activation function. wi ∼ N µ = 0, σ = 2 n bi = 0 • Recommended in practise. 47
  52. Deep learning in Practice: Pytorch Parameter Initilization Consider the previous

    model: model = torch.nn.Sequential( torch.nn.Linear(1,100), torch.nn.ReLU(), torch.nn.BatchNorm1d(100) torch.nn.Linear(100,50), torch.nn.ReLU(), torch.nn.BatchNorm1d(50) torch.nn.Linear(50,2)) To apply weight initilization to nn.linear module. def weights_init(m): if isinstance(m, nn.Linear): size = m.weight.size() n_out = size[0] n_in = size[1] variance = np.sqrt(2.0/(n_in + n_out) m.weight.data.normal_(0.0, variance) model.apply(weights_init) 48
  53. Outline Introduction to Deep learning Multilayer Perceptron Training Deep neural

    networks Deep learning in Practice Modern Deep learning Architecture Limitation and Research direction of deep neural networks 49
  54. Deep learning Architecture: Convolutional Neural Network Figure 8: CNN [credit:deeplearning.net]

    • Enhances the capabilities of MLP by inserting convolution layers. • Composed of many “filters”, which convolve, or slide across the data, and produce an activation at every slide position • Suitable for spatial data, object recognition and image analysis. 49
  55. Deep learning Architecture: Recurrent Neural Networks (RNN) RNN are neural

    networks with loops in them, allowing information to persist. • Can model a long time dimension and arbitrary sequence of events and inputs. • Suitable for sequenced data analysis: time-series, sentiment analysis, NLP, language translation, speech recognition etc. • Common type: LSTM and GRUs. 50
  56. Deep learning Architecture: Auto-enceoder Autoenceoder:A neural network where the input

    is the same as the output. Figure 9: credit:Arden Dertat • They compress the input into a lower-dimensional code and then reconstruct the output from this representation. • It is an unsupervised ML algorithm similar to PCA. • Several types exist: Denoising autoencoder, Sparse autoencoder. 51
  57. Deep learning Architecture: Auto-enceoder Autoencoder consists of components: encoder, code

    and decoder. • The encoder compresses the input and produces the code, • The decoder then reconstructs the input only using this code. Figure 10: credit:Arden Dertat 52
  58. Deep learning Architecture: Deep Generative models Idea:learn to understand data

    through generation → replicate the data distribution that you give it. • Can be used to generate Musics, Speach, Langauge, Image, Handwriting, Language • Suitable for unsupervised learning as they need lesser labelled data to train. Two types: 1 Autoregressive models: Deep NADE, PixelRNN, PixelCNN, WaveNet, ByteNet 2 Latent variable models: VAE, GAN. 53
  59. Outline Introduction to Deep learning Multilayer Perceptron Training Deep neural

    networks Deep learning in Practice Modern Deep learning Architecture Limitation and Research direction of deep neural networks 54
  60. Limitation • Very data hungry (eg. often millions of examples)

    • Computationally intensive to train and deploy (tractably requires GPUs) • Poor at representing uncertainty (how do you know what the model knows?) • Uninterpretable black boxes, difficult to trust • Difficult to optimize: non-convex, choice of architecture, learning parameters • Often require expert knowledge to design, fine tune architectures 54
  61. Research Direction • Transfer learning. • Unsepervised machine learning. •

    Computational efficiency. • Add more reasoning (uncertatinity) abilities ⇒ Bayesian Deep learning • Many applications which are under-explored especially in developing countries. 55
  62. Lab 3: Introduction to Deep learning Part 1: Feed-forward Neural

    Network (MLP): Objective: Build MLP classifier to recognize handwritten digits using the MNIST dataset. Part 2: Weight Initilization: Objective: Experiments with different initilization techniques (zero, xavier, kaiming) Part 3: Regularization: Objective: Experiments with different regulaization techniques (early stopping, dropout) 57
  63. References I • Deep learning for Artificial Intelligence master course:

    TelecomBCN Bercelona(winter 2017) • 6.S191 Introduction to Deep Learning: MIT 2018. • Deep learning Specilization by Andrew Ng: Coursera • Deep Learning by Russ Salakhutdinov: MLSS 2017 • Introductucion to Deep learning: CMU 2018 • Cs231n: Convolution Neural Network for Visual Recognition: Stanford 2018 • Deep learning in Pytorch, Francois Fleurent: EPFL 2018 • Advanced Machine Learning Specialization: Coursera 58