Lambton College Lecture 2: CBD-3335 Data Mining and Analysis PDF

Lambton College School of Computer Studies Lecture 2 CBD-3335 Data Mining and Analysis Artificial Neural Networks ◼The Brain ◼Brain vs. Computers ◼The Perceptron ◼Multilayer networks ◼Some Applications  Artificial Neural Networks ◼ Other te...

Lambton College School of Computer Studies Lecture 2 CBD-3335 Data Mining and Analysis Artificial Neural Networks ◼The Brain ◼Brain vs. Computers ◼The Perceptron ◼Multilayer networks ◼Some Applications  Artificial Neural Networks ◼ Other terms/names ◼ connectionist ◼ parallel distributed processing ◼ neural computation ◼ adaptive networks.. ◼ History ◼ 1943-McCulloch & Pitts are generally recognised as the designers of the first neural network ◼ 1949-First learning rule ◼ 1969-Minsky & Papert - perceptron limitation - Death of ANN ◼ 1980’s - Re-emergence of ANN - multi-layer networks 3  Brain and Machine The Brain – Pattern Recognition – Association – Complexity – Noise Tolerance The Machine – Calculation – Precision – Logic 4  The contrast in architecture The Von Neumann architecture uses a single processing unit; – Tens of millions of operations per second – Absolute arithmetic precision The brain uses many slow unreliable processors acting in parallel 5 Features of the Brain Ten billion (1010) neurons On average, several thousand connections Hundreds of operations per second Die off frequently (never replaced) Compensates for problems by massive parallelism 6  The biological inspiration The brain has been extensively studied by scientists. Vast complexity prevents all but rudimentary understanding. Even the behaviour of an individual neuron is extremely complex 7 The biological inspiration Single “percepts” distributed among many neurons Localized parts of the brain are responsible for certain well-defined functions (e.g. vision, motion). 8 The Structure of Neurons 9  The Structure of Neurons A neuron has a cell body, a branching input structure (the dendrIte) and a branching output structure (the axOn) Axons connect to dendrites via synapses. Electro-chemical signals are propagated from the dendritic input, through the cell body, and down the axon to other neurons 10  The Structure of Neurons A neuron only fires if its input signal exceeds a certain amount (the threshold) in a short time period. Synapses vary in strength – Good connections allowing a large signal – Slight connections allow only a weak signal. 11 12 The Artificial Neuron (Perceptron) ao+1 wj0 a1 wj1 wj2 a2 Sj f(Sj) Xj wjn an 13  A Simple Model of a Neuron (Perceptron) y1 w1j y2 w2j y3 w3j  O yi wij Each neuron has a threshold value Each neuron has weighted inputs from other neurons The input signals form a weighted sum If the activation level exceeds the threshold, the neuron “fires” 14  An Artificial Neuron y1 w1j y2 w2j y3 w3j  f(x) O yi wij Each hidden or output neuron has weighted input connections from each of the units in the preceding layer. The unit performs a weighted sum of its inputs, and subtracts its threshold value, to give its activation level. Activation level is passed through a sigmoid activation function to determine output. 15 Supervised Learning ◼ Training and test data sets ◼ Training set; input & target 16 Perceptron Training 1 if  wi xi >t Output={ i=0 0 otherwise ◼ Linear threshold is used. ◼ W - weight value ◼ t - threshold value 17 18  Simple network 1 if  wixi >t AND with a Biased input { output= i=0 0 otherwise -1 W1 = 1.5 X W2 = 1 t = 0.0 W3 = 1 Y 19  Learning algorithm While epoch produces an error Present network with next inputs from epoch Error = T – O If Error 0 then Wj = Wj + LR * Ij * Error End If End While 20  Learning algorithm Epoch : Presentation of the entire training set to the neural network. In the case of the AND function an epoch consists of four sets of inputs being presented to the network (i.e. [0,0], [0,1], [1,0], [1,1]) Error: The error value is the amount by which the value output by the network differs from the target value. For example, if we required the network to output 0 and it output a 1, then Error = -1 21  Learning algorithm Target Value, T : When we are training a network we not only present it with the input but also with a value that we require the network to produce. For example, if we present the network with [1,1] for the AND function the target value will be 1 Output , O : The output value from the neuron Ij : Inputs being presented to the neuron Wj : Weight from input neuron (Ij) to the output neuron LR : The learning rate. This dictates how quickly the network converges. It is set by a matter of experimentation. It is typically 0.1 22 Training Perceptrons For AND -1 A B Output W1 = ? 00 0 01 0 x t = 0.0 W2 = ? 10 0 11 1 W3 = ? y What are the weight values? Initialize with random weight values 23  Training Perceptrons For AND -1 A B Output W1 = 0.3 00 0 01 0 x t = 0.0 W2 = 0.5 10 0 11 1 W3 =-0.4 y I1 I2 I3 Summation Output -1 0 0 (-1*0.3) + (0*0.5) + (0*-0.4) = -0.3 0 -1 0 1 (-1*0.3) + (0*0.5) + (1*-0.4) = -0.7 0 -1 1 0 (-1*0.3) + (1*0.5) + (0*-0.4) = 0.2 1 -1 1 1 (-1*0.3) + (1*0.5) + (1*-0.4) = -0.2 0 24 Learning in Neural Networks ◼ Learn values of weights from I/O pairs ◼ Start with random weights ◼ Load training example’s input ◼ Observe computed input ◼ Modify weights to reduce difference ◼ Iterate over all training examples ◼ Terminate when weights stop changing OR when error is very small 25  Decision boundaries In simple cases, divide feature space by drawing a hyperplane across it. Known as a decision boundary. Discriminant function: returns different values on opposite sides. (straight line) Problems which can be thus classified are linearly separable. 26 Linear Separability X1 A A A B Decision A Boundary A B B B A B A B B X2 B 27  Rugby players & Ballet dancers 2 Rugby ? Height (m) Ballet? 1 50 120 Weight (Kg) 28  Hyperplane partitions A single Perceptron (i.e. output unit) with connections from each input can perform, and learn, a linear separation. Perceptrons have a step function activation. 29 Hyperplane partitions An extra layer models a convex hull – “An area with no dents in it” – Perceptron models, but can’t learn – Sigmoid function learning of convex hulls – Two layers add convex hulls together – Sufficient to classify anything “sane”. In theory, further layers add nothing In practice, extra layers may be better 30  Types of Layers The input layer. – Introduces input values into the network. – No activation function or other processing. The hidden layer(s). – Perform classification of features – Two hidden layers are sufficient to solve any problem – Features imply more layers may be better The output layer. – Functionally just like the hidden layers – Outputs are passed on to the world outside the neural network. 31 Multilayer Perceptron (MLP) Output Values Output Layer Adjustable Weights Input Layer Input Signals (External Stimuli) 32  Different Non-Linearly Separable Problems Types of Exclusive-OR Classes with Most General Structure Decision Regions Problem Meshed regions Region Shapes Single-Layer Half Plane A B Bounded By B A Hyperplane B A Two-Layer Convex Open A B Or B A Closed Regions B A Three-Layer Arbitrary (Complexity A B B Limited by No. A of Nodes) B A 33  Activation functions Transforms neuron’s input into output. Features of activation functions: A squashing effect is required Prevents accelerating growth of activation levels through the network. Simple and easy to calculate 34 Activation functions 35  Standard activation functions The hard-limiting threshold function – Corresponds to the biological paradigm either fires or not Sigmoid functions ('S'-shaped curves) 1 – The logistic function f(x) = 1 + e -ax – The hyperbolic tangent (symmetrical) – Both functions have a simple differential – Only the shape is important 36  Training Algorithms Adjust neural network weights to map inputs to outputs. Use a set of sample patterns where the desired output (given the inputs presented) is known. The purpose is to learn to generalize – Recognize features which are common to good and bad exemplars 37  Training Algorithms Cascade neurons together Output from one layer is the input to the next Each layer has its own sets of weights HKUST 38  Training Algorithms Predictions are fed forward through the network to classify HKUST 39  Training Algorithms Predictions are fed forward through the network to classify HKUST 40  Training Algorithms Predictions are fed forward through the network to classify HKUST 41  Training Algorithms Predictions are fed forward through the network to classify HKUST 42  Training Algorithms Predictions are fed forward through the network to classify HKUST 43  Training Algorithms Predictions are fed forward through the network to classify HKUST 44  Regularization L1, L2 regularization (weight decay) Dropout Randomly turn off some neurons Allows individual neurons to independently be responsible for performance Dropout: A simple way to prevent neural networks from overfitting [Srivastava JMLR 2014] Adapted from Jia-bin Huang  Back-Propagation A training procedure which allows multi-layer feedforward Neural Networks to be trained; Can theoretically perform “any” input-output mapping; Can learn to solve linearly inseparable problems. 46 Back-Propagation 47 GRADIENT DESCENT GRADIENT DESCENT GRADIENT DESCENT 50  Gradient descent We’ll update the weights Move in direction opposite to gradient: L Time Learning rate Figure from Andrej Karpathy  Gradient descent in multi-layer nets How to update the weights at all layers? Answer: backpropagation of error from higher layers to lower layers Figure from Andrej Karpathy  Mini-batch gradient descent Rather than compute the gradient from the loss for all training examples, could only use some of the data for each gradient update We cycle through all the training examples multiple times; each time we’ve cycled through all of them once is called an ‘epoch’ Allows faster training (e.g. on GPUs), parallelization Figure from Andrej Karpathy  Backpropagation: Error For output units (e.g. identity output, least squares loss): For hidden units: Backprop formula:  Applications The properties of neural networks define where they are useful. – Can learn complex mappings from inputs to outputs, based solely on samples – Difficult to analyse: firm predictions about neural network behaviour difficult; Unsuitable for safety-critical applications. – Require limited understanding from trainer, who can be guided by heuristics. 55 Neural network for OCR ◼ feedforward network A B ◼ trained using Back- C propagation D E Hidden Layer Output Layer Input Layer 56 OCR for 8x10 characters 10 10 10 8 8 8 ◼ NN are able to generalise ◼ learning involves generating a partitioning of the input space ◼ for single layer network input space must be linearly separable ◼ what is the dimension of this input space? ◼ how many points in the input space? ◼ this network is binary(uses binary values) ◼ networks may also be continuous 57  Engine management The behaviour of a car engine is influenced by a large number of parameters – temperature at various points – fuel/air mixture – lubricant viscosity. Major companies have used neural networks to dynamically tune an engine depending on current settings. 58  ALVINN Drives 70 mph on a public highway 30 outputs for steering 30x32 weights 4 hidden into one out of units four hidden 30x32 pixels unit as inputs 59  Signature recognition Each person's signature is different. There are structural similarities which are difficult to quantify. One company has manufactured a machine which recognizes signatures to within a high level of accuracy. – Considers speed in addition to gross shape. – Makes forgery even more difficult. 60  Sonar target recognition Distinguish mines from rocks on sea-bed The neural network is provided with a large number of parameters which are extracted from the sonar signal. The training set consists of sets of signals from rocks and mines. 61  Stock market prediction “Technical trading” refers to trading based solely on known statistical parameters; e.g. previous price Neural networks have been used to attempt to predict changes in prices. Difficult to assess success since companies using these techniques are reluctant to disclose information. 62  Mortgage assessment Assess risk of lending to an individual. Difficult to decide on marginal cases. Neural networks have been trained to make decisions, based upon the opinions of expert underwriters. Neural network produced a 12% reduction in delinquencies compared with human experts. 63 Neural Network Problems Many Parameters to be set Overfitting long training times... 64 Parameter setting Number of layers Number of neurons too many neurons, require more training time Learning rate from experience, value should be small ~0.1 Momentum term.. 65 Over-fitting With sufficient nodes can classify any training set exactly May have poor generalisation ability. Cross-validation with some patterns – Typically 30% of training patterns – Validation set error is checked each epoch – Stop training if validation error goes up 66  Training time How many epochs of training? – Stop if the error fails to improve (has reached a minimum) – Stop if the rate of improvement drops below a certain level – Stop if the error reaches an acceptable level – Stop when a certain number of epochs have passed 67  References https://www.slideserve.com/phelan-cortez/artificial-neural- networks https://media.giphy.com/media/4LiMmbAcvgTQs/giphy.gif https://miro.medium.com/max/1400/1*E- 5K5rHxCRTPrSWF60XLWw.gif https://media.giphy.com/media/O9rcZVmRcEGqI/giphy.gif https://media.giphy.com/media/4LiMmbAcvgTQs/giphy.gif 68 69