Introduction Biologically Motivated Crude Model Backpropagation
|
|
- Dorcas Lorena Mason
- 5 years ago
- Views:
Transcription
1 Introduction Biologically Motivated Crude Model Backpropagation 1
2 McCulloch-Pitts Neurons In 1943 Warren S. McCulloch, a neuroscientist, and Walter Pitts, a logician, published A logical calculus of the ideas immanent in nervous activity in [1]. Gave a highly simplified computational model of a neuron. 2
3 Rosenblatt In 1957 Rosenblatt, a neurobiologist at Cornell, was researching vision in flies. The neural processing that occurred within the eye itself particularly intrigued Rosenblatt and formed the basis of his Perceptron neural network [2]. The Perceptron and other models showed great promise with many initial successes. 3
4 Minsky and Papert In 1969 Marvin Minsky and Seymour Papert published a book [3] in which they discussed some of the limitations of the Perceptron model. Showed that the perceptron could not solve non-linear problems, such as the simple XOR problem. The effect of these problems was to limit much of the funding available for research into artificial neural networks. As a result, ANN research went into hibernation. 4
5 Werbos In 1974, Paul J. Werbos, published his Harvard University Ph.D. thesis [4], which first described the process of training artificial neural networks through the backpropagation of errors. Its significance was not noticed until it was rediscovered in 1986 by Rumelhart, Hinton, and Williams. 5
6 Hopfield In 1982, John Hopfield s work [5] caused a resurgence in the field. Hopfield's approach was not simply to create models but to develop technologies that could be applied to real life problems. Several books and conferences followed and provided a forum for people within the field to discuss the topic. 6
7 Rumelhart, Hinton, and Williams In 1986, Rumelhart, Hinton, and Williams rediscovered the backpropagation error learning algorithm. Now, very popular. Steve Grossberg, Teuvo Kohonen, and Henry Klopf also created new models. 7
8 IEEE In 1987 the Institute of Electrical and Electronic Engineers (IEEE) first International Conference on Neural Networks drew more than a thousand attendees. Many other conferences on ANNs appeared. 8
9 Support Vector Machines In 1990s, artificial neural networks were overtaken in popularity in machine learning by support vector machines and other, much simpler methods, such as linear classifiers. Renewed interest in neural nets was sparked in the 2000s by the advent of deep learning. 9
10 Natural Neuron: Basic Purpose From Dendrites Axon Axon terminals Soma (Cell body) Photo by: Sebastian Kaulitzki Nucleus Synapse The basic purpose of a neuron is to receive incoming information (in the form of chemical and electrical signals) and, based upon that information, determines whether or not to send an electrical signal (action potential) to other neurons, muscles, or glands. Thus, a natural neuron is like an electrochemical signal receiver and transmitter (transceiver). When a neuron sends an action potential, this electrical signal travels to the end terminals of the neuron (synapse), where it triggers a release of chemicals called neurotransmitters. The neurotransmitters cross a short gap between cells (synapse) and, are input by the adjoining cell (other neuron, muscle, or gland.) 10
11 Neuron Structure: 4 Parts From Dendrites Axon Axon terminals Soma (Cell body) Photo by: Sebastian Kaulitzki Nucleus Synapse A typical human neuron has a cell body (soma), an array of input paths or wires to receive incoming signals (dendrites), a single output path wire (axon) that carries electrical signals away from the neuron toward other cells, and many axon terminals (synapses). The dendrites are specialized to receive signals and transmit them toward the cell body. The single long axon carries action potentials away from the cell body. The synaptic terminals (1000 s) form connections either with the dendrites of other neurons or with effector cells in muscles or glands. 11
12 Neuron Communications: 4 Steps From Dendrites Axon Axon terminals Soma (Cell body) Photo by: Sebastian Kaulitzki Nucleus Synapse 1. A neuron receives information from the external environment or from other neurons. Human brain may receive input from up to 100,000 other neurons. 2. The neuron integrates the information from all of its inputs and determines whether or not to send an output signal, depending on the strength of the summed input. This integration takes place both in time (the duration of the input and the time between inputs) and in space (across the surface of the neuron). 3. The neuron propagates the signal along its axon (several meters with rates up to 100 m/s). 4. Finally, the neuron converts this electrical signal to a chemical one and transmits it to other neurons, muscles, or glands. 12
13 Neuron Communications: Synapse From Dendrites Axon Axon terminals Soma (Cell body) Nucleus Synapse Photo by: Alila Once an electrical signal has arrived at the end of an axon, the synaptic terminals release a chemical messenger called a neurotransmitter, which relays the signal across the synapse to the next neuron or to the effector cell. The magnitude, density of release, and type of chemical of the neurotransmitter released is not well understood, but the receiver s response can be either excitatory or inhibitory, depending on the properties of the receptor. 13
14 Crude Computational Model [8] transmitter neurons each send their electrical activation level (spikes) to neuron through and along the axon. Assumption: the precise timings of the spikes do not matter, and that only the frequency of the firing communicates information. Furthermore, the frequency of firing is modeled by an activation output; the higher is the output (this is supposed to model) the higher is the frequency of firing. (a) Above: Cartoon drawing of neuron. (b) Below: a model [8]. Activation of neuron 1: Axon from Synapse neuron 1 Dendrite q Cell body Output axon : Activation function 14
15 Crude Computational Model [8] The electrical activation level enter the synapses located at the junction of the axon terminals of the sending neurons and the dendrites of the receiving neuron. The synapse acts to multiplicatively amplify or attenuate the activation level, through a weight : thus, represents the strength of the neurotransmitters, which are input to the receiving neuron. The activation level of neuron depends on the sum of the neurotransmitters values and an activation function : If this sum is strong enough (greater than the neuron s threshold ), neuron outputs an excitatory signal; otherwise it sends an inhibitory signal through the axon, to other neurons or receptor cells. (a) Above: Cartoon drawing of neuron. (b) Below: a model [8]. Activation of neuron 1: Axon from Synapse neuron 1 Dendrite q Cell body : Activation function Output axon 15
16 This model of a biological neuron is very crude, simplified, and coarse. For example, there are many different types of neurons, each with different properties; thus, suggesting a requirement for a different model for each different type of neuron. The dendrites in biological neurons perform complex nonlinear computations; we are not even modeling this. The synapses are not just a single weight, they re a complex non-linear dynamical system. The exact timing of the output spikes in many systems is known to be important, suggesting that the rate code approximation may not hold. Due to all these and many other simplifications, please avoid trying to draw serious analogies between any neural network model and real brains. The neural network model has been modified and fit to solve computational problems, without trying to faithfully exemplify the real brain. See this [6], or more recently this [7] if you are interested to learn more details about the physiology of actual neurons. 16
17 The is called bias because the summed inputs need to have at least the level of the threshold (a bias) to excite the neuron. Since we don t know exactly the value of the threshold, we can model it to be learned as a weight. q Cell body Output axon The threshold of the activation function now becomes 0. q Cell body : Activation function Output axon : Activation function 17
18 The activation level of neuron depends on the sum of the neurotransmitters values ; If this sum is strong enough (greater than the neuron s threshold ), neuron outputs an excitatory signal; otherwise it sends an inhibitory signal through the axon, to other neurons or receptor cells. The is called bias because the summed inputs need to have at least the level of the threshold (a bias) to excite the neuron. Since we don t know exactly what value the threshold should be, we can learn it as a weight. 18
19 Common Activation Functions Sigmoid (Logistic) Function Hyperbolic Tangent Function 19
20 Common Activation Functions Rectified Linear Unit (ReLU) Convergence Rates --- ReLU ReLu was found to greatly accelerate (e.g. a factor of 6 in Krizhevsky et al.) the convergence of stochastic gradient descent compared to the sigmoid/tanh functions. It is argued that this is due to its linear, non-saturating form. 20
21 21
22 22
23 Comparison with Logistic Regression The model of a single neuron with a Sigmoid activation function is exactly the same as the model of the Logistic Regression classifier: Single Neuron Logistic Classifier Single neuron activation output: Input from -neurons. Logistic Regression prediction output: Input from -feature input. 23
24 Single Neuron Nonlinear Problem As we have seen with the linear Logistic Regression model, a single neuron with linear inputs is incapable of solving non-linear problems. For example, it cannot solve the XOR (or XNOR) problem, with given linear Boolean inputs More complex; Noisy examples; Analog training set; but same problem. 24
25 , , , Layer Layer Layer, Units in Layer can be interpreted as representing higher order features ( and ) of the data to realize the desired function. 25
26 Generic Neural Network Model For Higher Order Non-linear Problems,, Layer Input Layer Hidden Layer Hidden Layer Hidden Layer Output The leftmost layer of the network is called the input layer, and the rightmost layer the output layer. The input units are called inputs, where j is the feature of the t th training example. There can be any number of units in any layer. The circles labeled +1 are called bias units (or threshold units) and correspond to the intercept term. The middle layers of nodes are called the hidden layers, because either the input, desired output, or both input and desired output values of the nodes in the layer are not observed in the training set. 26
27 , Layer Layer Layer Our example neural network has 3 input units (not counting the bias unit), 3 hidden units, and 1 output unit. The connections between nodes are called weights, but are labeled for uniformity. 27
28 We label layer as, so layer is the input layer, and layer the output layer. Let denote the number of layers in our network. In our example network,. Our neural network has parameters: denotes the parameter (or weight) going to unit in layer and coming from unit in layer. (Note the order of the indices.), is the bias associated with unit in layer. Layer Layer Layer is the activation (meaning output value) of unit in layer. For Layer, we use to denote where j is the feature of the t th training example. is the total weighted sum of inputs to unit in layer, including the bias term: 28
29 Forward Propagation of Activations, Layer Layer Layer, Forward propagation corresponds to computing the output of each neuron in each layer, except for the input layer. 29
30 Method of Learning the Parameters (Weights): Gradient Descent We will use gradient descent to learn the parameters from the training set 30
31 Type of Weight Update: Online Mode As discussed earlier, there are two methods for updating the weights: Batch mode Update each weight after taking into consideration the effect of all training examples, i.e., after one training cycle. Online mode Update each weight after each randomly chosen training example. Weights are updated times per each training cycle, where is the number of training examples in the training set. Aka: Stochastic gradient descent, since the different results will be achieved due to the random nature of presenting training examples. In the development of the update equations, we will use the online mode, initially, to simplify the derivation. 31
32 For numepochs (training cycles) { For each training example chosen at random ( ) {//do for all training examples: For each weight in each layer { } } } Each weight is updated after taking into consideration only one randomly chosen training example. Note that the term corrected. is like an error term; it gives the amount by which should be is the learning rate. 32
33 Linear regression: Logistic regression: Neural Network: Problem is that with many units connected in network, the cost function will likely be non-convex, independent of what cost function is used, and therefore there will be multiple minima. Therefore, we will consider the Euclidean cost function (as in Linear regression, initially, to simplify the derivation of the weight update equations, but will substitute the Logistic cost function later. 33
34 Output Layer Error Expression For a network that has (i.e., multiple) output units, the output error expression for a single training example is: For our example network that has a single output unit, the error expression for a single training example is: Layer Layer Layer, We want to calculate in order to change the value of the weight to minimize the error : according to the gradient descent algorithm. 34
35 Computing the Impact of Changes in Weight Has on, Layer Layer Layer 35
36 Computing the Impact of Changes in Weight Has on, Layer Layer Layer 36
37 Computing the Impact of Changes in Weight Has on, Layer Layer Layer 37
38 Comparing Results & Compacting 38
39 Computing the Impact of Changes in Weight Has on, Layer Layer Layer 39
40 Alternative Method to Compute, Layer Layer Layer 40
41 Computing the Impact of Changes in Weight Has on, Layer Layer Layer 41
42 Compact Forms of and 42
43 For each training example chosen at random ( ) { For each weight in layer { 1 1, } } Layer Layer Layer Each weight is updated after performing one forward propagation of activations using one randomly chosen training example. Note that the term may be interpreted as an error term; it gives the amount by which should be corrected. Implementation note: the forward pass will have computed and and is known. 43
44 Hidden Layer Error Expression For a network that has (i.e., multiple) units in the output layer, the error expression for a single training example is: For our example network that has a single output unit, the error expression for a single training example is:, Layer Layer Layer We want to calculate in order to change the value of the weight to minimize the error : according to the gradient descent algorithm. 44
45 Computing the Impact of Changes in Weight Has on Let s Backpropagate the error!, Layer Layer Layer Note that subscript of is 1 for, corresponding to the j th subscript in. 45
46 In Common Chain Rule Path Note: This expression is the same for all, since all chain rule paths for weights in Layer 1 have that segment in common., In common chain rule path. Layer Layer Layer 46
47 Computing the Impact of Changes in Weight Has on Note: Subscripts of and are 1 because is the weight connected to unit 1 in Layer 2., Layer Layer Layer Note that subscript of is 2 for, corresponding to the j th subscript in. 47
48 Generalize: Impact of Changes in Weight Has on Note: Subscripts of and are 1 because is the weight connected to unit 1 in Layer 2., Layer Layer Layer Note that subscript of is j for, corresponding to the subscript in. 48
49 Computing the Impact of Changes in Weight Has on Note: Subscripts of and are 2 because is the weight connected to unit 2 in Layer 2., Layer Layer Layer Note that subscript of is 1 for, corresponding to the j th subscript in. 49
50 Computing the Impact of Changes in Weight Has on Note: Subscripts of and are 2 because is the weight connected to unit 2 in Layer 2., Layer Layer Layer Note that subscript of is 2 for, corresponding to the j th subscript in. 50
51 Generalize: the Impact of Changes in Weight Has on, Layer Layer Layer 51
52 , Layer Layer Layer 52
53 Generalize: the Impact of Changes in Weight Has on, Layer Layer Layer 53
54 For this Network Configuration 1. Forward propagation to determine activations: Layer Layer Layer 2. Back propagation with gradient descent to update weights: For each training example chosen at random ( ) { For each weight in layer { } } For each weight in layer { }, 54
55 Using Alternative Cost Functions Note that the only partial derivative that depends on the cost function is: Furthermore, that derivative affects only: Layer Layer Layer, Therefore, to use an alternative cost function, only the first term updated. of needs to be 55
56 Using Logistic Regression Cost Function For instance, to use the Logistic Cost Function (i.e., instead of ), we need to substitute the with the. Recall Logistic Regression s cost function and its derivative: For a single training example (i.e., for online mode): 56
57 References [1] W. S. McCulloch and W. Pitts, "A logical calculus of the ideas immanent in nervous activity," Bulletin of Mathematical Biophysics, vol. 5, pp , [2] F. Rosenblatt, "The Perceptron--a perceiving and recognizing automaton," Cornell Aeronautical Laboratory, New York, NY, [3] M. Minsky and S. Papert, Perceptrons: An Introduction to Computational Geometry, Cambridge MA: The MIT Press, [4] P. J. Werbos, "Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences," PhD thesis, Harvard University, Harvard, [5] J. J. Hopfield, "Neural networks and physical systems with emergent collective computational properties," in Proceedings of the National Academy of Sciences of the USA, [6] M. L. London and M. Hausser, "Dendritic Computation," [Online]. Available: [Accessed 05 February 2016]. [7] N. Brunel, V. Hakim and M. J. Richardson, "Single neuron dynamics and computation," Current Opinion in Neurobiology, vol. 25, pp , [8] Stanford and F.-F. Li, "Stanford University CS231n: Convolutional Neural Networks for Visual Recognition," 1 January [Online]. Available: [Accessed 05 February 2016]. 57
COMP9444 Neural Networks and Deep Learning 2. Perceptrons. COMP9444 c Alan Blair, 2017
COMP9444 Neural Networks and Deep Learning 2. Perceptrons COMP9444 17s2 Perceptrons 1 Outline Neurons Biological and Artificial Perceptron Learning Linear Separability Multi-Layer Networks COMP9444 17s2
More informationNeural Networks. Chapter 18, Section 7. TB Artificial Intelligence. Slides from AIMA 1/ 21
Neural Networks Chapter 8, Section 7 TB Artificial Intelligence Slides from AIMA http://aima.cs.berkeley.edu / 2 Outline Brains Neural networks Perceptrons Multilayer perceptrons Applications of neural
More informationArtificial Neural Networks. Historical description
Artificial Neural Networks Historical description Victor G. Lopez 1 / 23 Artificial Neural Networks (ANN) An artificial neural network is a computational model that attempts to emulate the functions of
More informationNeural networks. Chapter 19, Sections 1 5 1
Neural networks Chapter 19, Sections 1 5 Chapter 19, Sections 1 5 1 Outline Brains Neural networks Perceptrons Multilayer perceptrons Applications of neural networks Chapter 19, Sections 1 5 2 Brains 10
More informationNeural networks. Chapter 20, Section 5 1
Neural networks Chapter 20, Section 5 Chapter 20, Section 5 Outline Brains Neural networks Perceptrons Multilayer perceptrons Applications of neural networks Chapter 20, Section 5 2 Brains 0 neurons of
More informationMachine Learning. Neural Networks. (slides from Domingos, Pardo, others)
Machine Learning Neural Networks (slides from Domingos, Pardo, others) For this week, Reading Chapter 4: Neural Networks (Mitchell, 1997) See Canvas For subsequent weeks: Scaling Learning Algorithms toward
More informationData Mining Part 5. Prediction
Data Mining Part 5. Prediction 5.5. Spring 2010 Instructor: Dr. Masoud Yaghini Outline How the Brain Works Artificial Neural Networks Simple Computing Elements Feed-Forward Networks Perceptrons (Single-layer,
More informationArtificial Neural Networks The Introduction
Artificial Neural Networks The Introduction 01001110 01100101 01110101 01110010 01101111 01101110 01101111 01110110 01100001 00100000 01110011 01101011 01110101 01110000 01101001 01101110 01100001 00100000
More informationLecture 4: Feed Forward Neural Networks
Lecture 4: Feed Forward Neural Networks Dr. Roman V Belavkin Middlesex University BIS4435 Biological neurons and the brain A Model of A Single Neuron Neurons as data-driven models Neural Networks Training
More informationArtificial Neural Network and Fuzzy Logic
Artificial Neural Network and Fuzzy Logic 1 Syllabus 2 Syllabus 3 Books 1. Artificial Neural Networks by B. Yagnanarayan, PHI - (Cover Topologies part of unit 1 and All part of Unit 2) 2. Neural Networks
More informationNeural networks. Chapter 20. Chapter 20 1
Neural networks Chapter 20 Chapter 20 1 Outline Brains Neural networks Perceptrons Multilayer networks Applications of neural networks Chapter 20 2 Brains 10 11 neurons of > 20 types, 10 14 synapses, 1ms
More informationMachine Learning. Neural Networks. (slides from Domingos, Pardo, others)
Machine Learning Neural Networks (slides from Domingos, Pardo, others) For this week, Reading Chapter 4: Neural Networks (Mitchell, 1997) See Canvas For subsequent weeks: Scaling Learning Algorithms toward
More informationLast update: October 26, Neural networks. CMSC 421: Section Dana Nau
Last update: October 26, 207 Neural networks CMSC 42: Section 8.7 Dana Nau Outline Applications of neural networks Brains Neural network units Perceptrons Multilayer perceptrons 2 Example Applications
More informationMachine Learning. Neural Networks. (slides from Domingos, Pardo, others)
Machine Learning Neural Networks (slides from Domingos, Pardo, others) Human Brain Neurons Input-Output Transformation Input Spikes Output Spike Spike (= a brief pulse) (Excitatory Post-Synaptic Potential)
More informationFeedforward Neural Nets and Backpropagation
Feedforward Neural Nets and Backpropagation Julie Nutini University of British Columbia MLRG September 28 th, 2016 1 / 23 Supervised Learning Roadmap Supervised Learning: Assume that we are given the features
More informationCourse 395: Machine Learning - Lectures
Course 395: Machine Learning - Lectures Lecture 1-2: Concept Learning (M. Pantic) Lecture 3-4: Decision Trees & CBC Intro (M. Pantic & S. Petridis) Lecture 5-6: Evaluating Hypotheses (S. Petridis) Lecture
More informationEEE 241: Linear Systems
EEE 4: Linear Systems Summary # 3: Introduction to artificial neural networks DISTRIBUTED REPRESENTATION An ANN consists of simple processing units communicating with each other. The basic elements of
More informationNeural Networks: Introduction
Neural Networks: Introduction Machine Learning Fall 2017 Based on slides and material from Geoffrey Hinton, Richard Socher, Dan Roth, Yoav Goldberg, Shai Shalev-Shwartz and Shai Ben-David, and others 1
More informationIntroduction To Artificial Neural Networks
Introduction To Artificial Neural Networks Machine Learning Supervised circle square circle square Unsupervised group these into two categories Supervised Machine Learning Supervised Machine Learning Supervised
More informationArtificial Neural Networks. Q550: Models in Cognitive Science Lecture 5
Artificial Neural Networks Q550: Models in Cognitive Science Lecture 5 "Intelligence is 10 million rules." --Doug Lenat The human brain has about 100 billion neurons. With an estimated average of one thousand
More informationARTIFICIAL INTELLIGENCE. Artificial Neural Networks
INFOB2KI 2017-2018 Utrecht University The Netherlands ARTIFICIAL INTELLIGENCE Artificial Neural Networks Lecturer: Silja Renooij These slides are part of the INFOB2KI Course Notes available from www.cs.uu.nl/docs/vakken/b2ki/schema.html
More informationLinear Regression, Neural Networks, etc.
Linear Regression, Neural Networks, etc. Gradient Descent Many machine learning problems can be cast as optimization problems Define a function that corresponds to learning error. (More on this later)
More informationMachine Learning. Neural Networks
Machine Learning Neural Networks Bryan Pardo, Northwestern University, Machine Learning EECS 349 Fall 2007 Biological Analogy Bryan Pardo, Northwestern University, Machine Learning EECS 349 Fall 2007 THE
More informationNeural Networks (Part 1) Goals for the lecture
Neural Networks (Part ) Mark Craven and David Page Computer Sciences 760 Spring 208 www.biostat.wisc.edu/~craven/cs760/ Some of the slides in these lectures have been adapted/borrowed from materials developed
More informationIntroduction to Artificial Neural Networks
Facultés Universitaires Notre-Dame de la Paix 27 March 2007 Outline 1 Introduction 2 Fundamentals Biological neuron Artificial neuron Artificial Neural Network Outline 3 Single-layer ANN Perceptron Adaline
More information2018 EE448, Big Data Mining, Lecture 5. (Part II) Weinan Zhang Shanghai Jiao Tong University
2018 EE448, Big Data Mining, Lecture 5 Supervised Learning (Part II) Weinan Zhang Shanghai Jiao Tong University http://wnzhang.net http://wnzhang.net/teaching/ee448/index.html Content of Supervised Learning
More informationMachine Learning and Data Mining. Multi-layer Perceptrons & Neural Networks: Basics. Prof. Alexander Ihler
+ Machine Learning and Data Mining Multi-layer Perceptrons & Neural Networks: Basics Prof. Alexander Ihler Linear Classifiers (Perceptrons) Linear Classifiers a linear classifier is a mapping which partitions
More information(Feed-Forward) Neural Networks Dr. Hajira Jabeen, Prof. Jens Lehmann
(Feed-Forward) Neural Networks 2016-12-06 Dr. Hajira Jabeen, Prof. Jens Lehmann Outline In the previous lectures we have learned about tensors and factorization methods. RESCAL is a bilinear model for
More informationArtificial Neural Networks
Artificial Neural Networks 鮑興國 Ph.D. National Taiwan University of Science and Technology Outline Perceptrons Gradient descent Multi-layer networks Backpropagation Hidden layer representations Examples
More informationLecture 7 Artificial neural networks: Supervised learning
Lecture 7 Artificial neural networks: Supervised learning Introduction, or how the brain works The neuron as a simple computing element The perceptron Multilayer neural networks Accelerated learning in
More informationArtifical Neural Networks
Neural Networks Artifical Neural Networks Neural Networks Biological Neural Networks.................................. Artificial Neural Networks................................... 3 ANN Structure...........................................
More informationIntroduction to Neural Networks
Introduction to Neural Networks What are (Artificial) Neural Networks? Models of the brain and nervous system Highly parallel Process information much more like the brain than a serial computer Learning
More informationSGD and Deep Learning
SGD and Deep Learning Subgradients Lets make the gradient cheating more formal. Recall that the gradient is the slope of the tangent. f(w 1 )+rf(w 1 ) (w w 1 ) Non differentiable case? w 1 Subgradients
More informationArtificial Neural Networks (ANN) Xiaogang Su, Ph.D. Department of Mathematical Science University of Texas at El Paso
Artificial Neural Networks (ANN) Xiaogang Su, Ph.D. Department of Mathematical Science University of Texas at El Paso xsu@utep.edu Fall, 2018 Outline Introduction A Brief History ANN Architecture Terminology
More informationEE04 804(B) Soft Computing Ver. 1.2 Class 2. Neural Networks - I Feb 23, Sasidharan Sreedharan
EE04 804(B) Soft Computing Ver. 1.2 Class 2. Neural Networks - I Feb 23, 2012 Sasidharan Sreedharan www.sasidharan.webs.com 3/1/2012 1 Syllabus Artificial Intelligence Systems- Neural Networks, fuzzy logic,
More informationSections 18.6 and 18.7 Artificial Neural Networks
Sections 18.6 and 18.7 Artificial Neural Networks CS4811 - Artificial Intelligence Nilufer Onder Department of Computer Science Michigan Technological University Outline The brain vs. artifical neural
More informationSPSS, University of Texas at Arlington. Topics in Machine Learning-EE 5359 Neural Networks
Topics in Machine Learning-EE 5359 Neural Networks 1 The Perceptron Output: A perceptron is a function that maps D-dimensional vectors to real numbers. For notational convenience, we add a zero-th dimension
More informationInstituto Tecnológico y de Estudios Superiores de Occidente Departamento de Electrónica, Sistemas e Informática. Introductory Notes on Neural Networks
Introductory Notes on Neural Networs Dr. José Ernesto Rayas Sánche April Introductory Notes on Neural Networs Dr. José Ernesto Rayas Sánche BIOLOGICAL NEURAL NETWORKS The brain can be seen as a highly
More informationArtificial Intelligence
Artificial Intelligence Jeff Clune Assistant Professor Evolving Artificial Intelligence Laboratory Announcements Be making progress on your projects! Three Types of Learning Unsupervised Supervised Reinforcement
More informationNeural Networks. Fundamentals of Neural Networks : Architectures, Algorithms and Applications. L, Fausett, 1994
Neural Networks Neural Networks Fundamentals of Neural Networks : Architectures, Algorithms and Applications. L, Fausett, 1994 An Introduction to Neural Networks (nd Ed). Morton, IM, 1995 Neural Networks
More informationArtificial Neural Network
Artificial Neural Network Contents 2 What is ANN? Biological Neuron Structure of Neuron Types of Neuron Models of Neuron Analogy with human NN Perceptron OCR Multilayer Neural Network Back propagation
More informationLearning and Memory in Neural Networks
Learning and Memory in Neural Networks Guy Billings, Neuroinformatics Doctoral Training Centre, The School of Informatics, The University of Edinburgh, UK. Neural networks consist of computational units
More informationCS 4700: Foundations of Artificial Intelligence
CS 4700: Foundations of Artificial Intelligence Prof. Bart Selman selman@cs.cornell.edu Machine Learning: Neural Networks R&N 18.7 Intro & perceptron learning 1 2 Neuron: How the brain works # neurons
More informationIntroduction to Neural Networks
Introduction to Neural Networks Philipp Koehn 4 April 205 Linear Models We used before weighted linear combination of feature values h j and weights λ j score(λ, d i ) = j λ j h j (d i ) Such models can
More informationClassification goals: Make 1 guess about the label (Top-1 error) Make 5 guesses about the label (Top-5 error) No Bounding Box
ImageNet Classification with Deep Convolutional Neural Networks Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton Motivation Classification goals: Make 1 guess about the label (Top-1 error) Make 5 guesses
More informationPattern Recognition and Machine Learning. Artificial Neural networks
Pattern Recognition and Machine Learning Jaes L. Crowley ENSIMAG 3 - MMIS Fall Seester 2017 Lessons 7 20 Dec 2017 Outline Artificial Neural networks Notation...2 Introduction...3 Key Equations... 3 Artificial
More informationArtificial Neural Networks
Artificial Neural Networks CPSC 533 Winter 2 Christian Jacob Neural Networks in the Context of AI Systems Neural Networks as Mediators between Symbolic AI and Statistical Methods 2 5.-NeuralNets-2.nb Neural
More informationSections 18.6 and 18.7 Artificial Neural Networks
Sections 18.6 and 18.7 Artificial Neural Networks CS4811 - Artificial Intelligence Nilufer Onder Department of Computer Science Michigan Technological University Outline The brain vs artifical neural networks
More informationDEEP LEARNING AND NEURAL NETWORKS: BACKGROUND AND HISTORY
DEEP LEARNING AND NEURAL NETWORKS: BACKGROUND AND HISTORY 1 On-line Resources http://neuralnetworksanddeeplearning.com/index.html Online book by Michael Nielsen http://matlabtricks.com/post-5/3x3-convolution-kernelswith-online-demo
More informationCS 4700: Foundations of Artificial Intelligence
CS 4700: Foundations of Artificial Intelligence Prof. Bart Selman selman@cs.cornell.edu Machine Learning: Neural Networks R&N 18.7 Intro & perceptron learning 1 2 Neuron: How the brain works # neurons
More informationNeural Networks. Nicholas Ruozzi University of Texas at Dallas
Neural Networks Nicholas Ruozzi University of Texas at Dallas Handwritten Digit Recognition Given a collection of handwritten digits and their corresponding labels, we d like to be able to correctly classify
More informationNONLINEAR CLASSIFICATION AND REGRESSION. J. Elder CSE 4404/5327 Introduction to Machine Learning and Pattern Recognition
NONLINEAR CLASSIFICATION AND REGRESSION Nonlinear Classification and Regression: Outline 2 Multi-Layer Perceptrons The Back-Propagation Learning Algorithm Generalized Linear Models Radial Basis Function
More informationMachine Learning (CSE 446): Neural Networks
Machine Learning (CSE 446): Neural Networks Noah Smith c 2017 University of Washington nasmith@cs.washington.edu November 6, 2017 1 / 22 Admin No Wednesday office hours for Noah; no lecture Friday. 2 /
More informationNeural Networks and Fuzzy Logic Rajendra Dept.of CSE ASCET
Unit-. Definition Neural network is a massively parallel distributed processing system, made of highly inter-connected neural computing elements that have the ability to learn and thereby acquire knowledge
More informationCS:4420 Artificial Intelligence
CS:4420 Artificial Intelligence Spring 2018 Neural Networks Cesare Tinelli The University of Iowa Copyright 2004 18, Cesare Tinelli and Stuart Russell a a These notes were originally developed by Stuart
More informationRevision: Neural Network
Revision: Neural Network Exercise 1 Tell whether each of the following statements is true or false by checking the appropriate box. Statement True False a) A perceptron is guaranteed to perfectly learn
More informationApprentissage, réseaux de neurones et modèles graphiques (RCP209) Neural Networks and Deep Learning
Apprentissage, réseaux de neurones et modèles graphiques (RCP209) Neural Networks and Deep Learning Nicolas Thome Prenom.Nom@cnam.fr http://cedric.cnam.fr/vertigo/cours/ml2/ Département Informatique Conservatoire
More informationMultilayer Perceptron
Outline Hong Chang Institute of Computing Technology, Chinese Academy of Sciences Machine Learning Methods (Fall 2012) Outline Outline I 1 Introduction 2 Single Perceptron 3 Boolean Function Learning 4
More informationNeural Networks. Learning and Computer Vision Prof. Olga Veksler CS9840. Lecture 10
CS9840 Learning and Computer Vision Prof. Olga Veksler Lecture 0 Neural Networks Many slides are from Andrew NG, Yann LeCun, Geoffry Hinton, Abin - Roozgard Outline Short Intro Perceptron ( layer NN) Multilayer
More informationIntroduction to Neural Networks
Introduction to Neural Networks Philipp Koehn 3 October 207 Linear Models We used before weighted linear combination of feature values h j and weights λ j score(λ, d i ) = j λ j h j (d i ) Such models
More information4. Multilayer Perceptrons
4. Multilayer Perceptrons This is a supervised error-correction learning algorithm. 1 4.1 Introduction A multilayer feedforward network consists of an input layer, one or more hidden layers, and an output
More informationIntroduction and Perceptron Learning
Artificial Neural Networks Introduction and Perceptron Learning CPSC 565 Winter 2003 Christian Jacob Department of Computer Science University of Calgary Canada CPSC 565 - Winter 2003 - Emergent Computing
More informationIntelligent Systems: Reasoning and Recognition. Artificial Neural Networks
Intelligent Systes: Reasoning and Recognition Jaes L. Crowley MOSIG M1 Winter Seester 2018 Lesson 7 1 March 2018 Outline Artificial Neural Networks Notation...2 Introduction...3 Key Equations... 3 Artificial
More information17 Neural Networks NEURAL NETWORKS. x XOR 1. x Jonathan Richard Shewchuk
94 Jonathan Richard Shewchuk 7 Neural Networks NEURAL NETWORKS Can do both classification & regression. [They tie together several ideas from the course: perceptrons, logistic regression, ensembles of
More informationSupervised (BPL) verses Hybrid (RBF) Learning. By: Shahed Shahir
Supervised (BPL) verses Hybrid (RBF) Learning By: Shahed Shahir 1 Outline I. Introduction II. Supervised Learning III. Hybrid Learning IV. BPL Verses RBF V. Supervised verses Hybrid learning VI. Conclusion
More informationPart 8: Neural Networks
METU Informatics Institute Min720 Pattern Classification ith Bio-Medical Applications Part 8: Neural Netors - INTRODUCTION: BIOLOGICAL VS. ARTIFICIAL Biological Neural Netors A Neuron: - A nerve cell as
More informationNeural Networks Lecturer: J. Matas Authors: J. Matas, B. Flach, O. Drbohlav
Neural Networks 30.11.2015 Lecturer: J. Matas Authors: J. Matas, B. Flach, O. Drbohlav 1 Talk Outline Perceptron Combining neurons to a network Neural network, processing input to an output Learning Cost
More informationComputational Intelligence Winter Term 2009/10
Computational Intelligence Winter Term 2009/10 Prof. Dr. Günter Rudolph Lehrstuhl für Algorithm Engineering (LS 11) Fakultät für Informatik TU Dortmund Plan for Today Organization (Lectures / Tutorials)
More informationNeural Networks Introduction
Neural Networks Introduction H.A Talebi Farzaneh Abdollahi Department of Electrical Engineering Amirkabir University of Technology Winter 2011 H. A. Talebi, Farzaneh Abdollahi Neural Networks 1/22 Biological
More informationDEVS Simulation of Spiking Neural Networks
DEVS Simulation of Spiking Neural Networks Rene Mayrhofer, Michael Affenzeller, Herbert Prähofer, Gerhard Höfer, Alexander Fried Institute of Systems Science Systems Theory and Information Technology Johannes
More informationHopfield Neural Network and Associative Memory. Typical Myelinated Vertebrate Motoneuron (Wikipedia) Topic 3 Polymers and Neurons Lecture 5
Hopfield Neural Network and Associative Memory Typical Myelinated Vertebrate Motoneuron (Wikipedia) PHY 411-506 Computational Physics 2 1 Wednesday, March 5 1906 Nobel Prize in Physiology or Medicine.
More informationNeural Networks. CSE 6363 Machine Learning Vassilis Athitsos Computer Science and Engineering Department University of Texas at Arlington
Neural Networks CSE 6363 Machine Learning Vassilis Athitsos Computer Science and Engineering Department University of Texas at Arlington 1 Perceptrons x 0 = 1 x 1 x 2 z = h w T x Output: z x D A perceptron
More informationCourse Overview. before 1943: Neurobiology. before 1943: Neurobiology. before 1943: Neurobiology. The Start of Artificial Neural Nets
Course Overview Introduction to Artificial Neural Networks Kristof Van Laerhoven kristof @ mis.tu-darmstadt.de http://www.mis.informatik.tu-darmstadt.de/kristof We will: Learn to interpret common ANN formulas
More informationComputational Intelligence
Plan for Today Computational Intelligence Winter Term 29/ Organization (Lectures / Tutorials) Overview CI Introduction to ANN McCulloch Pitts Neuron (MCP) Minsky / Papert Perceptron (MPP) Prof. Dr. Günter
More informationDeep Learning. Ali Ghodsi. University of Waterloo
University of Waterloo Deep learning attempts to learn representations of data with multiple levels of abstraction. Deep learning usually refers to a set of algorithms and computational models that are
More informationNeural Networks and Deep Learning
Neural Networks and Deep Learning Professor Ameet Talwalkar November 12, 2015 Professor Ameet Talwalkar Neural Networks and Deep Learning November 12, 2015 1 / 16 Outline 1 Review of last lecture AdaBoost
More informationArtificial Neural Networks. MGS Lecture 2
Artificial Neural Networks MGS 2018 - Lecture 2 OVERVIEW Biological Neural Networks Cell Topology: Input, Output, and Hidden Layers Functional description Cost functions Training ANNs Back-Propagation
More informationArtificial neural networks
Artificial neural networks Chapter 8, Section 7 Artificial Intelligence, spring 203, Peter Ljunglöf; based on AIMA Slides c Stuart Russel and Peter Norvig, 2004 Chapter 8, Section 7 Outline Brains Neural
More informationCISC 3250 Systems Neuroscience
CISC 3250 Systems Neuroscience Systems Neuroscience How the nervous system performs computations How groups of neurons work together to achieve intelligence Professor Daniel Leeds dleeds@fordham.edu JMH
More informationChapter 9: The Perceptron
Chapter 9: The Perceptron 9.1 INTRODUCTION At this point in the book, we have completed all of the exercises that we are going to do with the James program. These exercises have shown that distributed
More informationIn the Name of God. Lecture 9: ANN Architectures
In the Name of God Lecture 9: ANN Architectures Biological Neuron Organization of Levels in Brains Central Nervous sys Interregional circuits Local circuits Neurons Dendrite tree map into cerebral cortex,
More informationCSC321 Lecture 5: Multilayer Perceptrons
CSC321 Lecture 5: Multilayer Perceptrons Roger Grosse Roger Grosse CSC321 Lecture 5: Multilayer Perceptrons 1 / 21 Overview Recall the simple neuron-like unit: y output output bias i'th weight w 1 w2 w3
More informationCMSC 421: Neural Computation. Applications of Neural Networks
CMSC 42: Neural Computation definition synonyms neural networks artificial neural networks neural modeling connectionist models parallel distributed processing AI perspective Applications of Neural Networks
More informationMultilayer Neural Networks. (sometimes called Multilayer Perceptrons or MLPs)
Multilayer Neural Networks (sometimes called Multilayer Perceptrons or MLPs) Linear separability Hyperplane In 2D: w x + w 2 x 2 + w 0 = 0 Feature x 2 = w w 2 x w 0 w 2 Feature 2 A perceptron can separate
More informationFundamentals of Neural Networks
Fundamentals of Neural Networks : Soft Computing Course Lecture 7 14, notes, slides www.myreaders.info/, RC Chakraborty, e-mail rcchak@gmail.com, Aug. 10, 2010 http://www.myreaders.info/html/soft_computing.html
More informationGrundlagen der Künstlichen Intelligenz
Grundlagen der Künstlichen Intelligenz Neural networks Daniel Hennes 21.01.2018 (WS 2017/18) University Stuttgart - IPVS - Machine Learning & Robotics 1 Today Logistic regression Neural networks Perceptron
More information22c145-Fall 01: Neural Networks. Neural Networks. Readings: Chapter 19 of Russell & Norvig. Cesare Tinelli 1
Neural Networks Readings: Chapter 19 of Russell & Norvig. Cesare Tinelli 1 Brains as Computational Devices Brains advantages with respect to digital computers: Massively parallel Fault-tolerant Reliable
More informationLinear discriminant functions
Andrea Passerini passerini@disi.unitn.it Machine Learning Discriminative learning Discriminative vs generative Generative learning assumes knowledge of the distribution governing the data Discriminative
More informationFundamentals of Neural Network
Chapter 3 Fundamentals of Neural Network One of the main challenge in actual times researching, is the construction of AI (Articial Intelligence) systems. These systems could be understood as any physical
More informationDeep Neural Networks
Deep Neural Networks DT2118 Speech and Speaker Recognition Giampiero Salvi KTH/CSC/TMH giampi@kth.se VT 2015 1 / 45 Outline State-to-Output Probability Model Artificial Neural Networks Perceptron Multi
More informationCS 6501: Deep Learning for Computer Graphics. Basics of Neural Networks. Connelly Barnes
CS 6501: Deep Learning for Computer Graphics Basics of Neural Networks Connelly Barnes Overview Simple neural networks Perceptron Feedforward neural networks Multilayer perceptron and properties Autoencoders
More informationComputational Intelligence
Plan for Today Computational Intelligence Winter Term 207/8 Organization (Lectures / Tutorials) Overview CI Introduction to ANN McCulloch Pitts Neuron (MCP) Minsky / Papert Perceptron (MPP) Prof. Dr. Günter
More informationMultilayer Perceptron Tutorial
Multilayer Perceptron Tutorial Leonardo Noriega School of Computing Staffordshire University Beaconside Staffordshire ST18 0DG email: l.a.noriega@staffs.ac.uk November 17, 2005 1 Introduction to Neural
More information2015 Todd Neller. A.I.M.A. text figures 1995 Prentice Hall. Used by permission. Neural Networks. Todd W. Neller
2015 Todd Neller. A.I.M.A. text figures 1995 Prentice Hall. Used by permission. Neural Networks Todd W. Neller Machine Learning Learning is such an important part of what we consider "intelligence" that
More informationCN2 1: Introduction. Paul Gribble. Sep 10,
CN2 1: Introduction Paul Gribble http://gribblelab.org Sep 10, 2012 Administrivia Class meets Mondays, 2:00pm - 3:30pm and Thursdays, 11:30am - 1:00pm, in NSC 245A Contact me with any questions or to set
More informationCOMP 551 Applied Machine Learning Lecture 14: Neural Networks
COMP 551 Applied Machine Learning Lecture 14: Neural Networks Instructor: Ryan Lowe (ryan.lowe@mail.mcgill.ca) Slides mostly by: Class web page: www.cs.mcgill.ca/~hvanho2/comp551 Unless otherwise noted,
More informationArtificial Neural Networks" and Nonparametric Methods" CMPSCI 383 Nov 17, 2011!
Artificial Neural Networks" and Nonparametric Methods" CMPSCI 383 Nov 17, 2011! 1 Todayʼs lecture" How the brain works (!)! Artificial neural networks! Perceptrons! Multilayer feed-forward networks! Error
More informationNeural Networks biological neuron artificial neuron 1
Neural Networks biological neuron artificial neuron 1 A two-layer neural network Output layer (activation represents classification) Weighted connections Hidden layer ( internal representation ) Input
More informationIntroduction to Convolutional Neural Networks (CNNs)
Introduction to Convolutional Neural Networks (CNNs) nojunk@snu.ac.kr http://mipal.snu.ac.kr Department of Transdisciplinary Studies Seoul National University, Korea Jan. 2016 Many slides are from Fei-Fei
More informationLecture 4: Perceptrons and Multilayer Perceptrons
Lecture 4: Perceptrons and Multilayer Perceptrons Cognitive Systems II - Machine Learning SS 2005 Part I: Basic Approaches of Concept Learning Perceptrons, Artificial Neuronal Networks Lecture 4: Perceptrons
More information