CS B553: A LGORITHMS FOR O PTIMIZATION AND L EARNING Gradient descent.

Slides:



Advertisements
Similar presentations
Introduction to Optimization Anjela Govan North Carolina State University SAMSI NDHS Undergraduate workshop 2006.
Advertisements

CS B553: A LGORITHMS FOR O PTIMIZATION AND L EARNING Univariate optimization.
Optimization : The min and max of a function
CHAPTER 2 D IRECT M ETHODS FOR S TOCHASTIC S EARCH Organization of chapter in ISSO –Introductory material –Random search methods Attributes of random search.
Machine learning optimization
Steepest Decent and Conjugate Gradients (CG). Solving of the linear equation system.
Geometry Optimization Pertemuan VI. Geometry Optimization Backgrounds Real molecules vibrate thermally about their equilibrium structures. Finding minimum.
Nonlinear Optimization for Optimal Control
Design Optimization School of Engineering University of Bradford 1 Numerical optimization techniques Unconstrained multi-parameter optimization techniques.
Optimization via Search CPSC 315 – Programming Studio Spring 2009 Project 2, Lecture 4 Adapted from slides of Yoonsuck Choe.
Search and Optimization Methods Based in part on Chapter 8 of Hand, Manilla, & Smyth David Madigan.
Optimization Methods One-Dimensional Unconstrained Optimization
September 23, 2010Neural Networks Lecture 6: Perceptron Learning 1 Refresher: Perceptron Training Algorithm Algorithm Perceptron; Start with a randomly.
Math for CSTutorial 5-61 Tutorial 5 Function Optimization. Line Search. Taylor Series for R n Steepest Descent.
Advanced Topics in Optimization
Why Function Optimization ?
CS 4700: Foundations of Artificial Intelligence
Optimization Methods One-Dimensional Unconstrained Optimization
Optimization via Search CPSC 315 – Programming Studio Spring 2008 Project 2, Lecture 4 Adapted from slides of Yoonsuck Choe.
What is Optimization? Optimization is the mathematical discipline which is concerned with finding the maxima and minima of functions, possibly subject.
9 1 Performance Optimization. 9 2 Basic Optimization Algorithm p k - Search Direction  k - Learning Rate or.
Computational Optimization
UNCONSTRAINED MULTIVARIABLE
Learning with large datasets Machine Learning Large scale machine learning.
Nonlinear programming Unconstrained optimization techniques.
Local Search Algorithms This lecture topic Chapter Next lecture topic Chapter 5 (Please read lecture topic material before and after each lecture.
CS 478 – Tools for Machine Learning and Data Mining Backpropagation.
B EYOND C LASSICAL S EARCH Instructor: Kris Hauser 1.
1 Unconstrained Optimization Objective: Find minimum of F(X) where X is a vector of design variables We may know lower and upper bounds for optimum No.
EECS 274 Computer Vision Segmentation by Clustering II.
Course: Logic Programming and Constraints
Quasi-Newton Methods of Optimization Lecture 2. General Algorithm n A Baseline Scenario Algorithm U (Model algorithm for n- dimensional unconstrained.
Particle Swarm Optimization by Dr. Shubhajit Roy Chowdhury Centre for VLSI and Embedded Systems Technology, IIIT Hyderabad.
CS621 : Artificial Intelligence
Chapter 10 Minimization or Maximization of Functions.
1 Chapter 6 General Strategy for Gradient methods (1) Calculate a search direction (2) Select a step length in that direction to reduce f(x) Steepest Descent.
Exam 1 Oct 3, closed book Place ITE 119, Time:12:30-1:45pm
L25 Numerical Methods part 5 Project Questions Homework Review Tips and Tricks Summary 1.
Chapter 2-OPTIMIZATION G.Anuradha. Contents Derivative-based Optimization –Descent Methods –The Method of Steepest Descent –Classical Newton’s Method.
METHOD OF STEEPEST DESCENT ELE Adaptive Signal Processing1 Week 5.
CS B553: A LGORITHMS FOR O PTIMIZATION AND L EARNING Constrained optimization.
INTRO TO OPTIMIZATION MATH-415 Numerical Analysis 1.
Bab 5 Classification: Alternative Techniques Part 4 Artificial Neural Networks Based Classifer.
Geometry Optimizations Lecture CompChem 2 Chemistry 347 Hope College.
숭실대 전기공학과 C ontrol I nformation P rocess L ab 김경진.
Numerical Methods for Inverse Kinematics Kris Hauser ECE 383 / ME 442.
Optimization via Search
Fall 2004 Backpropagation CS478 - Machine Learning.
Heuristic Optimization Methods Optimization as Search
Real Neurons Cell structures Cell body Dendrites Axon
(Fractions / Surds /Indices)
Discussion #29 – Images II
KMeans Clustering on Hadoop Fall 2013 Elke A. Rundensteiner
Function Notation “f of x” Input = x Output = f(x) = y.
METHOD OF STEEPEST DESCENT
Introduction to Scientific Computing II
Introduction to Scientific Computing II
Introduction to Scientific Computing II
Introduction to Scientific Computing II
Multi-layer perceptron
More on Search: A* and Optimization
3.8 Newton’s Method How do you find a root of the following function without a graphing calculator? This is what Newton did.
Least squares linear classifier
More on HW 2 (due Jan 26) Again, it must be in Python 2.7.
Introduction to Scientific Computing II
More on HW 2 (due Jan 26) Again, it must be in Python 2.7.
Neural Network Training
Gradient/Concavity 2.5 Geometrical Application of Calculus
Performance Optimization
Steepest Descent Optimization
Presentation transcript:

CS B553: A LGORITHMS FOR O PTIMIZATION AND L EARNING Gradient descent

K EY C ONCEPTS Gradient descent Line search Convergence rates depend on scaling Variants: discrete analogues, coordinate descent Random restarts

Line search: pick step size to lead to decrease in function value

(Use your favorite univariate optimization method)  f(x-  f(x)) 

G RADIENT D ESCENT P SEUDOCODE Input: f, starting value x 1, termination tolerances For t=1,2,…,maxIters: Compute the search direction d t = -  f ( x t ) If || d t ||< ε g then: return “Converged to critical point”, output x t Find  t so that f ( x t +  t d t ) < f ( x t ) using line search If ||  t d t ||< ε x then: return “Converged in x ”, output x t Let x t +1 = x t +  t d t Return “Max number of iterations reached”, output x maxIters

R ELATED M ETHODS Steepest descent (discrete) Coordinate descent

Many local minima: good initialization, or random restarts