Machine learning for crystal structure prediction

Size: px
Start display at page:

Download "Machine learning for crystal structure prediction"

Transcription

1 Machine learning for crystal structure prediction Fei Qi Computational Material Design Laboratory Moscow Institute of Physics and Technology Dolgoprudny, Moscow December 12, 2014

2 Motivation Introduction to Machine Learning Machine Learning based Property Prediction Results of KRR based Property Prediction Discussions and Conclusions

3 Complexity of Crystal Structure Prediction Crystal structure prediction (CSP) To find stable crystal structures To find a structure with some expected physical/chemical properties. Computation of properties First principle calculation (FPC) is with very high complexity. In CSP, a large number of structures are populated and evaluated using FPC The most time consuming step in CSP.

4 Evolutionary Crystal Structure Prediction USPEX 1 provides a smart evolutionary approach to populate structures. But the number of structures to be evaluated is still large. Workflow of USPEX 1. Initial structure population 2. Applying evolutionary operators to structures 3. Structure relaxation and property computation 4. New generation 1 C. W. Glass, A. R. Oganov, and N. Hansen, Computer Physics Communications 175, 713 (2006).

5 Evolutionary Crystal Structure Prediction USPEX 1 provides a smart evolutionary approach to populate structures. But the number of structures to be evaluated is still large. Workflow of USPEX 1. Initial structure population 2. Applying evolutionary operators to structures 3. Structure relaxation and property computation 4. New generation Can we speed-up the process? 1 C. W. Glass, A. R. Oganov, and N. Hansen, Computer Physics Communications 175, 713 (2006).

6 Evolutionary Crystal Structure Prediction USPEX 1 provides a smart evolutionary approach to populate structures. But the number of structures to be evaluated is still large. Workflow of USPEX 1. Initial structure population 2. Applying evolutionary operators to structures 3. Structure relaxation and property computation 4. New generation Can we speed-up the process? To provide an efficient alternative to step 3. Or to filter structures after step 2. 1 C. W. Glass, A. R. Oganov, and N. Hansen, Computer Physics Communications 175, 713 (2006).

7 Introduction to Machine Learning Data Algorithm Model f (x)

8 Supervised Learning for Property Prediction Data Algorithm Model f (x)

9 Supervised Learning for Property Prediction Data Algorithm Model f ( ) =

10 y To Speed-Up Property Computation with Machine Learning Machine learning (ML) can learn models from data. Training is to build the model according to known data. Prediction applies model on unknown data. ML can accelerate property computation The training procedure is very efficient (seconds vs hours) The prediction procedure is extremely fast (ms vs hours) A typical ML approach is regression analysis Model y =2x +1 Linear Regression Kernel Ridge Regression Model y =sin(x +π/7) +1 Kernel Regression y =2x x x

11 Applying machine learning applied to CSP Property Prediction Model P = f (X) Feature Extraction X

12 Feature extraction Feature vector Coordinates of atoms Bonding length Bonding angles...

13 Feature extraction Feature vector Coordinates of atoms Bonding length Bonding angles... Challenges in CSP Performance depends on feature vector Need to be designed for different compounds Leads to complex models Returns back to FPC Difficult to train

14 Feature extraction: an exemplar-based approach Exemplar feature vector Similarities to reference structures K(, ) K(, ) K(, ).

15 Motivation Formulation Method Results Summary Why examples makes sense Intuitively, similar structure are with similar property. Exemplar-based approaches have successful applications in the image processing community, including image inpainting2, super-resolution, etc. Energy/property learning shares some properties with image processing High-dimensional space Globally non-linear but locally linear property. 2 C. Guillemot and O. Le Meur, Image Inpainting: Overview and Recent Advances, IEEE Signal Processing Magazine, 31(1): , Jan

16 The Kernel Ridge Regression Model The property of an unknown structure is predicted using P pred (f) = N i=1 α ik (f i, f), where f and f i are fingerprints of the unknown and known structures. The training minimize the following goal function N i=1 Pi prediction Pi train 2 + λ N The optimization has a closed form solution α = (K + λi) 1 P train, where the kernel matrix K is K ij = K ( f i, f j ). i=1 α2 i

17 The Fingerprint Function The Gaussian kernel is used, which is ( / K(x, t) exp x t 2 2 2σ 2). The fingerprint provides a consistent representation is V uc δ(r R ij) f AB (R) = 4π N A N B Rij 2 1. The corresponding descriptor is obtained by discretization f AB = [f AB (0), f AB ( ), f AB (2 ),, f AB (n )] T, where A and B can be all combinations of A, B fulfill A B. A i B j

18 Sparsity and feature/sample selection Sparsity If we have a large population of referencing structures, only a small portion of them are useful. Vector form of the KRR model Kα P train λ α 2 2 Applying sparsity into the model Kα P train λ α 0 Difference between atoms can be considered. Using atomic fingerprint Number of training samples is less than that of atoms. Assumes non-uniform contributions to a property

19 Evaluation Method Performance measurement Person correlation coefficient K-fold cross-validation 1 2 K Divide the training set to K subsets To select 1 subset for testing Other K 1 subsets are used to train the model

20 Evaluation Method Performance measurement Person correlation coefficient K-fold cross-validation 1 2 K Divide the training set to K subsets To select 1 subset for testing Other K 1 subsets are used to train the model To guarantee for generalization capability

21 Prediction results of properties of BN Predicted properties of BN: density (left), hardness (upper right), and bandgap (bottom right).

22 Prediction results of average enthalpy Prediction results of average enthalpy of: left) BN with variable combination and right) Ag2Se with 24 atoms at 30GPa.

23 Predicting atomization energy of organic molecules The QM7 dataset contains 7165 molecules The median absolute error is 7.6 kcal/mol Error of reference 4 is 7.6 kcal/mol. Results on the QM7 dataset M. Rupp, A. Tkatchenko, K.-R. Mller, and O. A. von Lilienfeld, Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning, Physics Review Letters, 108(5), Jan

24 To boost the prediction with USPEX Pearson Coefficients 100% 95% 90% 85% 80% 75% ML accuracy w.r.t. generation on average enthalpy of BN Density Bandgap Hardness Enthalpy_avg 70% 0% 20% 40% 60% 80% 100% Percentage of training samples Training samples are increased according to USPEX s evolutionary generations. The rest samples are used in testing. The generalization capability increases along with the generation.

25 Summary Conclusions Exemplar-based feature vector is introduced for CSP with machine learning. Kernel ridge regression (KRR) can be used in the prediction of material properties. Sparsity and feature selection are briefly introduced. With machine learning, structures (populated by USPEX) can be pre-filtered (or evaluated very efficiently with tolerable errors).

26 Thanks for your attention. Acknowledgments Artem R. Oganov (State University, New York & MIPT) Qiang Zhu (State University, New York) Guangrui Qian (State University, New York) Qingfeng Zeng (Northwestern Polytechnique University, China) Haiyang Niu, Qinggao Wang and Xiaohu Yu (MIPT) This work is funded by the grant of the Government of the Russian Federation NO. 14.A

Supervised Machine Learning: Learning SVMs and Deep Learning. Klaus-Robert Müller!!et al.!!

Supervised Machine Learning: Learning SVMs and Deep Learning. Klaus-Robert Müller!!et al.!! Supervised Machine Learning: Learning SVMs and Deep Learning Klaus-Robert Müller!!et al.!! Today s Tutorial Machine Learning introduction: ingredients for ML Kernel Methods and Deep networks with explaining

More information

Supplementary Online Materials: Formation of Stoichiometric CsF n Compounds

Supplementary Online Materials: Formation of Stoichiometric CsF n Compounds 1 2 3 4 5 6 7 8 9 1 11 12 13 Supplementary Online Materials: Formation of Stoichiometric CsF n Compounds Qiang Zhu, 1, a) Artem R. Oganov, 1, 2, 3 and Qingfeng Zeng 4 1) Department of Geosciences, Stony

More information

Machine Learning and Sparsity. Klaus-Robert Müller!!et al.!!

Machine Learning and Sparsity. Klaus-Robert Müller!!et al.!! Machine Learning and Sparsity Klaus-Robert Müller!!et al.!! Today s Talk sensing, sparse models and generalization interpretabilty and sparse methods explaining for nonlinear methods Sparse Models & Generalization?

More information

Quantum Mechanics, Chemical Space, And Machine Learning Or

Quantum Mechanics, Chemical Space, And Machine Learning Or Quantum Mechanics, Chemical Space, And Machine Learning Or B Anatole von Lilienfeld Institute of Physical Chemistry and National Center for Computational Design and Discovery of Novel Materials (MARVEL),

More information

Less is More: Computational Regularization by Subsampling

Less is More: Computational Regularization by Subsampling Less is More: Computational Regularization by Subsampling Lorenzo Rosasco University of Genova - Istituto Italiano di Tecnologia Massachusetts Institute of Technology lcsl.mit.edu joint work with Alessandro

More information

Prediction of double gene knockout measurements

Prediction of double gene knockout measurements Prediction of double gene knockout measurements Sofia Kyriazopoulou-Panagiotopoulou sofiakp@stanford.edu December 12, 2008 Abstract One way to get an insight into the potential interaction between a pair

More information

Localized Coulomb Descriptors for the Gaussian Approximation Potential

Localized Coulomb Descriptors for the Gaussian Approximation Potential Localized Coulomb Descriptors for the Gaussian Approximation Potential James Barker 1,2, Johannes Bulin 1, Jan Hamaekers 1, and Sonja Mathias 1 1 Fraunhofer-Institut für Algorithmen und Wissenschaftliches

More information

Less is More: Computational Regularization by Subsampling

Less is More: Computational Regularization by Subsampling Less is More: Computational Regularization by Subsampling Lorenzo Rosasco University of Genova - Istituto Italiano di Tecnologia Massachusetts Institute of Technology lcsl.mit.edu joint work with Alessandro

More information

Electricity Demand Forecasting using Multi-Task Learning

Electricity Demand Forecasting using Multi-Task Learning Electricity Demand Forecasting using Multi-Task Learning Jean-Baptiste Fiot, Francesco Dinuzzo Dublin Machine Learning Meetup - July 2017 1 / 32 Outline 1 Introduction 2 Problem Formulation 3 Kernels 4

More information

Bond type restricted radial distribution functions for accurate machine learning prediction of atomization energies

Bond type restricted radial distribution functions for accurate machine learning prediction of atomization energies Bond type restricted radial distribution functions for accurate machine learning prediction of atomization energies Mykhaylo Krykunov, Tom K. Woo Department of Chemistry and Biomolecular Science University

More information

6.036 midterm review. Wednesday, March 18, 15

6.036 midterm review. Wednesday, March 18, 15 6.036 midterm review 1 Topics covered supervised learning labels available unsupervised learning no labels available semi-supervised learning some labels available - what algorithms have you learned that

More information

Gaussian Processes. Le Song. Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012

Gaussian Processes. Le Song. Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012 Gaussian Processes Le Song Machine Learning II: Advanced Topics CSE 8803ML, Spring 01 Pictorial view of embedding distribution Transform the entire distribution to expected features Feature space Feature

More information

Inclusion of Machine Learning Kernel Ridge Regression. Potential Energy Surfaces in On-the-Fly Nonadiabatic

Inclusion of Machine Learning Kernel Ridge Regression. Potential Energy Surfaces in On-the-Fly Nonadiabatic Supporting information for Inclusion of Machine Learning Kernel Ridge Regression Potential Energy Surfaces in On-the-Fly Nonadiabatic Molecular Dynamics Simulation Deping Hu, Yu Xie, Xusong Li, Lingyue

More information

Introduction to Support Vector Machines

Introduction to Support Vector Machines Introduction to Support Vector Machines Shivani Agarwal Support Vector Machines (SVMs) Algorithm for learning linear classifiers Motivated by idea of maximizing margin Efficient extension to non-linear

More information

Tutorial on Interpreting and Explaining Deep Models in Computer Vision

Tutorial on Interpreting and Explaining Deep Models in Computer Vision Tutorial on Interpreting and Explaining Deep Models in Computer Vision Wojciech Samek (Fraunhofer HHI) Grégoire Montavon (TU Berlin) Klaus-Robert Müller (TU Berlin) 08:30-09:15 Introduction KRM 09:15-10:00

More information

Machine Learning for Atomic Forces in a Crystalline Solid: Transferability to Various Temperatures

Machine Learning for Atomic Forces in a Crystalline Solid: Transferability to Various Temperatures arxiv:1608.07374v1 [cond-mat.mtrl-sci] 26 Aug 2016 Machine Learning for Atomic Forces in a Crystalline Solid: Transferability to Various Temperatures Teppei Suzuki, Ryo Tamura, Tsuyoshi Miyazaki August

More information

Kernel-based Machine Learning for Virtual Screening

Kernel-based Machine Learning for Virtual Screening Kernel-based Machine Learning for Virtual Screening Dipl.-Inf. Matthias Rupp Beilstein Endowed Chair for Chemoinformatics Johann Wolfgang Goethe-University Frankfurt am Main, Germany 2008-04-11, Helmholtz

More information

Introduction to Machine Learning

Introduction to Machine Learning Introduction to Machine Learning Kernel Methods Varun Chandola Computer Science & Engineering State University of New York at Buffalo Buffalo, NY, USA chandola@buffalo.edu Chandola@UB CSE 474/574 1 / 21

More information

Modeling Data with Linear Combinations of Basis Functions. Read Chapter 3 in the text by Bishop

Modeling Data with Linear Combinations of Basis Functions. Read Chapter 3 in the text by Bishop Modeling Data with Linear Combinations of Basis Functions Read Chapter 3 in the text by Bishop A Type of Supervised Learning Problem We want to model data (x 1, t 1 ),..., (x N, t N ), where x i is a vector

More information

Predicting Variable Stoichiometric Compounds. Qiang Zhu Department of Geosciences, Stony Brook University

Predicting Variable Stoichiometric Compounds. Qiang Zhu Department of Geosciences, Stony Brook University Predicting Variable Stoichiometric Compounds Qiang Zhu Department of Geosciences, Stony Brook University USPEX: Computational Materials Design Crystal Structure Prediction Dimension 0: Nano-particle; 1:

More information

Kernel methods, kernel SVM and ridge regression

Kernel methods, kernel SVM and ridge regression Kernel methods, kernel SVM and ridge regression Le Song Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012 Collaborative Filtering 2 Collaborative Filtering R: rating matrix; U: user factor;

More information

CIS 520: Machine Learning Oct 09, Kernel Methods

CIS 520: Machine Learning Oct 09, Kernel Methods CIS 520: Machine Learning Oct 09, 207 Kernel Methods Lecturer: Shivani Agarwal Disclaimer: These notes are designed to be a supplement to the lecture They may or may not cover all the material discussed

More information

Few-shot learning with KRR

Few-shot learning with KRR Few-shot learning with KRR Prudencio Tossou Groupe de Recherche en Apprentissage Automatique Départment d informatique et de génie logiciel Université Laval April 6, 2018 Prudencio Tossou (UL) Few-shot

More information

Multiple Kernel Learning

Multiple Kernel Learning CS 678A Course Project Vivek Gupta, 1 Anurendra Kumar 2 Sup: Prof. Harish Karnick 1 1 Department of Computer Science and Engineering 2 Department of Electrical Engineering Indian Institute of Technology,

More information

Interaction Potentials in Molecules and Non-Local Information in Chemical Space

Interaction Potentials in Molecules and Non-Local Information in Chemical Space Interaction Potentials in Molecules and Non-Local Information in Chemical Space Katja Hansen 1, Franziska Biegler 2, O. Anatole von Lilienfeld 3,4, Klaus-Robert Müller 2,5, and Alexandre Tkatchenko 1 1

More information

Collaborative topic models: motivations cont

Collaborative topic models: motivations cont Collaborative topic models: motivations cont Two topics: machine learning social network analysis Two people: " boy Two articles: article A! girl article B Preferences: The boy likes A and B --- no problem.

More information

Outline Introduction OLS Design of experiments Regression. Metamodeling. ME598/494 Lecture. Max Yi Ren

Outline Introduction OLS Design of experiments Regression. Metamodeling. ME598/494 Lecture. Max Yi Ren 1 / 34 Metamodeling ME598/494 Lecture Max Yi Ren Department of Mechanical Engineering, Arizona State University March 1, 2015 2 / 34 1. preliminaries 1.1 motivation 1.2 ordinary least square 1.3 information

More information

Computer Vision Group Prof. Daniel Cremers. 10a. Markov Chain Monte Carlo

Computer Vision Group Prof. Daniel Cremers. 10a. Markov Chain Monte Carlo Group Prof. Daniel Cremers 10a. Markov Chain Monte Carlo Markov Chain Monte Carlo In high-dimensional spaces, rejection sampling and importance sampling are very inefficient An alternative is Markov Chain

More information

Nearest Neighbor. Machine Learning CSE546 Kevin Jamieson University of Washington. October 26, Kevin Jamieson 2

Nearest Neighbor. Machine Learning CSE546 Kevin Jamieson University of Washington. October 26, Kevin Jamieson 2 Nearest Neighbor Machine Learning CSE546 Kevin Jamieson University of Washington October 26, 2017 2017 Kevin Jamieson 2 Some data, Bayes Classifier Training data: True label: +1 True label: -1 Optimal

More information

Supplementary Materials for

Supplementary Materials for advances.sciencemag.org/cgi/content/full/3/5/e1603015/dc1 Supplementary Materials for Machine learning of accurate energy-conserving molecular force fields This PDF file includes: Stefan Chmiela, Alexandre

More information

Machine learning for pervasive systems Classification in high-dimensional spaces

Machine learning for pervasive systems Classification in high-dimensional spaces Machine learning for pervasive systems Classification in high-dimensional spaces Department of Communications and Networking Aalto University, School of Electrical Engineering stephan.sigg@aalto.fi Version

More information

Big Data Analytics. Special Topics for Computer Science CSE CSE Feb 24

Big Data Analytics. Special Topics for Computer Science CSE CSE Feb 24 Big Data Analytics Special Topics for Computer Science CSE 4095-001 CSE 5095-005 Feb 24 Fei Wang Associate Professor Department of Computer Science and Engineering fei_wang@uconn.edu Prediction III Goal

More information

Kernel Logistic Regression and the Import Vector Machine

Kernel Logistic Regression and the Import Vector Machine Kernel Logistic Regression and the Import Vector Machine Ji Zhu and Trevor Hastie Journal of Computational and Graphical Statistics, 2005 Presented by Mingtao Ding Duke University December 8, 2011 Mingtao

More information

Modeling Electronic Quantum Transport with Machine Learning

Modeling Electronic Quantum Transport with Machine Learning Modeling Electronic Quantum Transport with Machine Learning Alejandro Lopez-Bezanilla 1 and O. Anatole von Lilienfeld 2,3 1 Materials Science Division, Argonne National Laboratory, 9700 S. Cass Avenue,

More information

Relevance Vector Machines for Earthquake Response Spectra

Relevance Vector Machines for Earthquake Response Spectra 2012 2011 American American Transactions Transactions on on Engineering Engineering & Applied Applied Sciences Sciences. American Transactions on Engineering & Applied Sciences http://tuengr.com/ateas

More information

STOCHASTIC INFORMATION GRADIENT ALGORITHM BASED ON MAXIMUM ENTROPY DENSITY ESTIMATION. Badong Chen, Yu Zhu, Jinchun Hu and Ming Zhang

STOCHASTIC INFORMATION GRADIENT ALGORITHM BASED ON MAXIMUM ENTROPY DENSITY ESTIMATION. Badong Chen, Yu Zhu, Jinchun Hu and Ming Zhang ICIC Express Letters ICIC International c 2009 ISSN 1881-803X Volume 3, Number 3, September 2009 pp. 1 6 STOCHASTIC INFORMATION GRADIENT ALGORITHM BASED ON MAXIMUM ENTROPY DENSITY ESTIMATION Badong Chen,

More information

Statistical Machine Learning from Data

Statistical Machine Learning from Data Samy Bengio Statistical Machine Learning from Data 1 Statistical Machine Learning from Data Support Vector Machines Samy Bengio IDIAP Research Institute, Martigny, Switzerland, and Ecole Polytechnique

More information

Outline. Motivation. Mapping the input space to the feature space Calculating the dot product in the feature space

Outline. Motivation. Mapping the input space to the feature space Calculating the dot product in the feature space to The The A s s in to Fabio A. González Ph.D. Depto. de Ing. de Sistemas e Industrial Universidad Nacional de Colombia, Bogotá April 2, 2009 to The The A s s in 1 Motivation Outline 2 The Mapping the

More information

Introduction to SVM and RVM

Introduction to SVM and RVM Introduction to SVM and RVM Machine Learning Seminar HUS HVL UIB Yushu Li, UIB Overview Support vector machine SVM First introduced by Vapnik, et al. 1992 Several literature and wide applications Relevance

More information

Machine learning comes from Bayesian decision theory in statistics. There we want to minimize the expected value of the loss function.

Machine learning comes from Bayesian decision theory in statistics. There we want to minimize the expected value of the loss function. Bayesian learning: Machine learning comes from Bayesian decision theory in statistics. There we want to minimize the expected value of the loss function. Let y be the true label and y be the predicted

More information

Computational Materials Discovery: New Methods and Results. Artem R. Oganov (ARO)

Computational Materials Discovery: New Methods and Results. Artem R. Oganov (ARO) Computational Materials Discovery: New Methods and Results Artem R. Oganov (ARO) Moscow Institute of Physics and Technology, Dolgoprudny, Russia State University of New York, Stony Brook, USA Northwestern

More information

Machine Learning 2: Nonlinear Regression

Machine Learning 2: Nonlinear Regression 15-884 Machine Learning : Nonlinear Regression J. Zico Kolter September 17, 01 1 Non-linear regression Peak Hourly Demand (GW).5 0 0 40 60 80 100 High temperature / peak demand observations for all days

More information

The evolution from MLE to MAP to Bayesian Learning

The evolution from MLE to MAP to Bayesian Learning The evolution from MLE to MAP to Bayesian Learning Zhe Li January 13, 2015 1 The evolution from MLE to MAP to Bayesian Based on the linear regression, we illustrate the evolution from Maximum Loglikelihood

More information

Structured Prediction

Structured Prediction Structured Prediction Ningshan Zhang Advanced Machine Learning, Spring 2016 Outline Ensemble Methods for Structured Prediction[1] On-line learning Boosting AGeneralizedKernelApproachtoStructuredOutputLearning[2]

More information

Non-Bayesian Classifiers Part II: Linear Discriminants and Support Vector Machines

Non-Bayesian Classifiers Part II: Linear Discriminants and Support Vector Machines Non-Bayesian Classifiers Part II: Linear Discriminants and Support Vector Machines Selim Aksoy Department of Computer Engineering Bilkent University saksoy@cs.bilkent.edu.tr CS 551, Fall 2018 CS 551, Fall

More information

Predicting Protein Interactions with Motifs

Predicting Protein Interactions with Motifs Predicting Protein Interactions with Motifs Jessica Long Chetan Sharma Lekan Wang December 12, 2008 1 Background Proteins are essential to almost all living organisms. They are comprised of a long, tangled

More information

Final Overview. Introduction to ML. Marek Petrik 4/25/2017

Final Overview. Introduction to ML. Marek Petrik 4/25/2017 Final Overview Introduction to ML Marek Petrik 4/25/2017 This Course: Introduction to Machine Learning Build a foundation for practice and research in ML Basic machine learning concepts: max likelihood,

More information

Introduction to Machine Learning Midterm, Tues April 8

Introduction to Machine Learning Midterm, Tues April 8 Introduction to Machine Learning 10-701 Midterm, Tues April 8 [1 point] Name: Andrew ID: Instructions: You are allowed a (two-sided) sheet of notes. Exam ends at 2:45pm Take a deep breath and don t spend

More information

... SPARROW. SPARse approximation Weighted regression. Pardis Noorzad. Department of Computer Engineering and IT Amirkabir University of Technology

... SPARROW. SPARse approximation Weighted regression. Pardis Noorzad. Department of Computer Engineering and IT Amirkabir University of Technology ..... SPARROW SPARse approximation Weighted regression Pardis Noorzad Department of Computer Engineering and IT Amirkabir University of Technology Université de Montréal March 12, 2012 SPARROW 1/47 .....

More information

Regularized Least Squares

Regularized Least Squares Regularized Least Squares Ryan M. Rifkin Google, Inc. 2008 Basics: Data Data points S = {(X 1, Y 1 ),...,(X n, Y n )}. We let X simultaneously refer to the set {X 1,...,X n } and to the n by d matrix whose

More information

Jeff Howbert Introduction to Machine Learning Winter

Jeff Howbert Introduction to Machine Learning Winter Classification / Regression Support Vector Machines Jeff Howbert Introduction to Machine Learning Winter 2012 1 Topics SVM classifiers for linearly separable classes SVM classifiers for non-linearly separable

More information

Statistical learning theory, Support vector machines, and Bioinformatics

Statistical learning theory, Support vector machines, and Bioinformatics 1 Statistical learning theory, Support vector machines, and Bioinformatics Jean-Philippe.Vert@mines.org Ecole des Mines de Paris Computational Biology group ENS Paris, november 25, 2003. 2 Overview 1.

More information

Machine Learning. Regularization and Feature Selection. Fabio Vandin November 13, 2017

Machine Learning. Regularization and Feature Selection. Fabio Vandin November 13, 2017 Machine Learning Regularization and Feature Selection Fabio Vandin November 13, 2017 1 Learning Model A: learning algorithm for a machine learning task S: m i.i.d. pairs z i = (x i, y i ), i = 1,..., m,

More information

First-Principles Determination of the Structure of Magnesium Borohydride

First-Principles Determination of the Structure of Magnesium Borohydride First-Principles Determination of the Structure of Magnesium Borohydride Xiang-Feng Zhou, 1, 2, Artem R. Oganov, 2, 3 Guang-Rui Qian, 2 and Qiang Zhu 2 1 School of Physics and Key Laboratory of Weak-Light

More information

Sparse regression. Optimization-Based Data Analysis. Carlos Fernandez-Granda

Sparse regression. Optimization-Based Data Analysis.   Carlos Fernandez-Granda Sparse regression Optimization-Based Data Analysis http://www.cims.nyu.edu/~cfgranda/pages/obda_spring16 Carlos Fernandez-Granda 3/28/2016 Regression Least-squares regression Example: Global warming Logistic

More information

PREDICTING SOLAR GENERATION FROM WEATHER FORECASTS. Chenlin Wu Yuhan Lou

PREDICTING SOLAR GENERATION FROM WEATHER FORECASTS. Chenlin Wu Yuhan Lou PREDICTING SOLAR GENERATION FROM WEATHER FORECASTS Chenlin Wu Yuhan Lou Background Smart grid: increasing the contribution of renewable in grid energy Solar generation: intermittent and nondispatchable

More information

Kernel Machines. Pradeep Ravikumar Co-instructor: Manuela Veloso. Machine Learning

Kernel Machines. Pradeep Ravikumar Co-instructor: Manuela Veloso. Machine Learning Kernel Machines Pradeep Ravikumar Co-instructor: Manuela Veloso Machine Learning 10-701 SVM linearly separable case n training points (x 1,, x n ) d features x j is a d-dimensional vector Primal problem:

More information

Linear Regression. CSL603 - Fall 2017 Narayanan C Krishnan

Linear Regression. CSL603 - Fall 2017 Narayanan C Krishnan Linear Regression CSL603 - Fall 2017 Narayanan C Krishnan ckn@iitrpr.ac.in Outline Univariate regression Multivariate regression Probabilistic view of regression Loss functions Bias-Variance analysis Regularization

More information

Machine learning - HT Basis Expansion, Regularization, Validation

Machine learning - HT Basis Expansion, Regularization, Validation Machine learning - HT 016 4. Basis Expansion, Regularization, Validation Varun Kanade University of Oxford Feburary 03, 016 Outline Introduce basis function to go beyond linear regression Understanding

More information

Linear Regression. CSL465/603 - Fall 2016 Narayanan C Krishnan

Linear Regression. CSL465/603 - Fall 2016 Narayanan C Krishnan Linear Regression CSL465/603 - Fall 2016 Narayanan C Krishnan ckn@iitrpr.ac.in Outline Univariate regression Multivariate regression Probabilistic view of regression Loss functions Bias-Variance analysis

More information

Signal Recovery from Permuted Observations

Signal Recovery from Permuted Observations EE381V Course Project Signal Recovery from Permuted Observations 1 Problem Shanshan Wu (sw33323) May 8th, 2015 We start with the following problem: let s R n be an unknown n-dimensional real-valued signal,

More information

Gaussian processes and bayesian optimization Stanisław Jastrzębski. kudkudak.github.io kudkudak

Gaussian processes and bayesian optimization Stanisław Jastrzębski. kudkudak.github.io kudkudak Gaussian processes and bayesian optimization Stanisław Jastrzębski kudkudak.github.io kudkudak Plan Goal: talk about modern hyperparameter optimization algorithms Bayes reminder: equivalent linear regression

More information

Gaussian Processes. 1 What problems can be solved by Gaussian Processes?

Gaussian Processes. 1 What problems can be solved by Gaussian Processes? Statistical Techniques in Robotics (16-831, F1) Lecture#19 (Wednesday November 16) Gaussian Processes Lecturer: Drew Bagnell Scribe:Yamuna Krishnamurthy 1 1 What problems can be solved by Gaussian Processes?

More information

Holdout and Cross-Validation Methods Overfitting Avoidance

Holdout and Cross-Validation Methods Overfitting Avoidance Holdout and Cross-Validation Methods Overfitting Avoidance Decision Trees Reduce error pruning Cost-complexity pruning Neural Networks Early stopping Adjusting Regularizers via Cross-Validation Nearest

More information

Computing regularization paths for learning multiple kernels

Computing regularization paths for learning multiple kernels Computing regularization paths for learning multiple kernels Francis Bach Romain Thibaux Michael Jordan Computer Science, UC Berkeley December, 24 Code available at www.cs.berkeley.edu/~fbach Computing

More information

Regularized Least Squares

Regularized Least Squares Regularized Least Squares Charlie Frogner 1 MIT 2011 1 Slides mostly stolen from Ryan Rifkin (Google). Summary In RLS, the Tikhonov minimization problem boils down to solving a linear system (and this

More information

Evaluation. Andrea Passerini Machine Learning. Evaluation

Evaluation. Andrea Passerini Machine Learning. Evaluation Andrea Passerini passerini@disi.unitn.it Machine Learning Basic concepts requires to define performance measures to be optimized Performance of learning algorithms cannot be evaluated on entire domain

More information

Machine Learning! in just a few minutes. Jan Peters Gerhard Neumann

Machine Learning! in just a few minutes. Jan Peters Gerhard Neumann Machine Learning! in just a few minutes Jan Peters Gerhard Neumann 1 Purpose of this Lecture Foundations of machine learning tools for robotics We focus on regression methods and general principles Often

More information

LINEAR CLASSIFICATION, PERCEPTRON, LOGISTIC REGRESSION, SVC, NAÏVE BAYES. Supervised Learning

LINEAR CLASSIFICATION, PERCEPTRON, LOGISTIC REGRESSION, SVC, NAÏVE BAYES. Supervised Learning LINEAR CLASSIFICATION, PERCEPTRON, LOGISTIC REGRESSION, SVC, NAÏVE BAYES Supervised Learning Linear vs non linear classifiers In K-NN we saw an example of a non-linear classifier: the decision boundary

More information

ML (cont.): SUPPORT VECTOR MACHINES

ML (cont.): SUPPORT VECTOR MACHINES ML (cont.): SUPPORT VECTOR MACHINES CS540 Bryan R Gibson University of Wisconsin-Madison Slides adapted from those used by Prof. Jerry Zhu, CS540-1 1 / 40 Support Vector Machines (SVMs) The No-Math Version

More information

The Kernel Trick, Gram Matrices, and Feature Extraction. CS6787 Lecture 4 Fall 2017

The Kernel Trick, Gram Matrices, and Feature Extraction. CS6787 Lecture 4 Fall 2017 The Kernel Trick, Gram Matrices, and Feature Extraction CS6787 Lecture 4 Fall 2017 Momentum for Principle Component Analysis CS6787 Lecture 3.1 Fall 2017 Principle Component Analysis Setting: find the

More information

Kernel Methods. Konstantin Tretyakov MTAT Machine Learning

Kernel Methods. Konstantin Tretyakov MTAT Machine Learning Kernel Methods Konstantin Tretyakov (kt@ut.ee) MTAT.03.227 Machine Learning So far Supervised machine learning Linear models Non-linear models Unsupervised machine learning Generic scaffolding So far Supervised

More information

New from Ulf in Berzerkeley

New from Ulf in Berzerkeley New from Ulf in Berzerkeley from crystallization to statistics of density fluctuations Ulf Rørbæk Pedersen Department of Chemistry, University of California, Berkeley, USA Roskilde, December 16th, 21 Ulf

More information

Evaluation requires to define performance measures to be optimized

Evaluation requires to define performance measures to be optimized Evaluation Basic concepts Evaluation requires to define performance measures to be optimized Performance of learning algorithms cannot be evaluated on entire domain (generalization error) approximation

More information

Leaving The Span Manfred K. Warmuth and Vishy Vishwanathan

Leaving The Span Manfred K. Warmuth and Vishy Vishwanathan Leaving The Span Manfred K. Warmuth and Vishy Vishwanathan UCSC and NICTA Talk at NYAS Conference, 10-27-06 Thanks to Dima and Jun 1 Let s keep it simple Linear Regression 10 8 6 4 2 0 2 4 6 8 8 6 4 2

More information

Overview of Statistical Tools. Statistical Inference. Bayesian Framework. Modeling. Very simple case. Things are usually more complicated

Overview of Statistical Tools. Statistical Inference. Bayesian Framework. Modeling. Very simple case. Things are usually more complicated Fall 3 Computer Vision Overview of Statistical Tools Statistical Inference Haibin Ling Observation inference Decision Prior knowledge http://www.dabi.temple.edu/~hbling/teaching/3f_5543/index.html Bayesian

More information

Midterm. Introduction to Machine Learning. CS 189 Spring You have 1 hour 20 minutes for the exam.

Midterm. Introduction to Machine Learning. CS 189 Spring You have 1 hour 20 minutes for the exam. CS 189 Spring 2013 Introduction to Machine Learning Midterm You have 1 hour 20 minutes for the exam. The exam is closed book, closed notes except your one-page crib sheet. Please use non-programmable calculators

More information

Machine Learning, Midterm Exam

Machine Learning, Midterm Exam 10-601 Machine Learning, Midterm Exam Instructors: Tom Mitchell, Ziv Bar-Joseph Wednesday 12 th December, 2012 There are 9 questions, for a total of 100 points. This exam has 20 pages, make sure you have

More information

A Bayesian Perspective on Residential Demand Response Using Smart Meter Data

A Bayesian Perspective on Residential Demand Response Using Smart Meter Data A Bayesian Perspective on Residential Demand Response Using Smart Meter Data Datong-Paul Zhou, Maximilian Balandat, and Claire Tomlin University of California, Berkeley [datong.zhou, balandat, tomlin]@eecs.berkeley.edu

More information

Real Estate Price Prediction with Regression and Classification CS 229 Autumn 2016 Project Final Report

Real Estate Price Prediction with Regression and Classification CS 229 Autumn 2016 Project Final Report Real Estate Price Prediction with Regression and Classification CS 229 Autumn 2016 Project Final Report Hujia Yu, Jiafu Wu [hujiay, jiafuwu]@stanford.edu 1. Introduction Housing prices are an important

More information

Machine Learning: Assignment 1

Machine Learning: Assignment 1 10-701 Machine Learning: Assignment 1 Due on Februrary 0, 014 at 1 noon Barnabas Poczos, Aarti Singh Instructions: Failure to follow these directions may result in loss of points. Your solutions for this

More information

https://goo.gl/kfxweg KYOTO UNIVERSITY Statistical Machine Learning Theory Sparsity Hisashi Kashima kashima@i.kyoto-u.ac.jp DEPARTMENT OF INTELLIGENCE SCIENCE AND TECHNOLOGY 1 KYOTO UNIVERSITY Topics:

More information

Kernel Methods. Konstantin Tretyakov MTAT Machine Learning

Kernel Methods. Konstantin Tretyakov MTAT Machine Learning Kernel Methods Konstantin Tretyakov (kt@ut.ee) MTAT.03.227 Machine Learning So far Supervised machine learning Linear models Least squares regression, SVR Fisher s discriminant, Perceptron, Logistic model,

More information

Patterns, Correlations, and Causality in Big Data of Materials: Analytics for Novel Materials Discovery

Patterns, Correlations, and Causality in Big Data of Materials: Analytics for Novel Materials Discovery Patterns, Correlations, and Causality in Big Data of Materials: Analytics for Novel Materials Discovery From the periodic table of the elements to a chart (a map) of materials: Organize materials according

More information

Gaussian Approximation Potentials: a brief tutorial introduction

Gaussian Approximation Potentials: a brief tutorial introduction Gaussian Approximation Potentials: a brief tutorial introduction Albert P. Bartók and Gábor Csányi Engineering Laboratory, University of Cambridge, Trumpington Street, Cambridge, CB2 1PZ, United Kingdom

More information

Crystal Structure Prediction and Its Application in Earth and Materials Sciences

Crystal Structure Prediction and Its Application in Earth and Materials Sciences Top Curr Chem (2014) 345: 223 256 DOI: 10.1007/128_2013_508 # Springer-Verlag Berlin Heidelberg 2014 Published online: 15 February 2014 Crystal Structure Prediction and Its Application in Earth and Materials

More information

WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials

WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials One of the obvious choices to obtain rotational and translational invariance in the descriptor is the use

More information

Basis Expansion and Nonlinear SVM. Kai Yu

Basis Expansion and Nonlinear SVM. Kai Yu Basis Expansion and Nonlinear SVM Kai Yu Linear Classifiers f(x) =w > x + b z(x) = sign(f(x)) Help to learn more general cases, e.g., nonlinear models 8/7/12 2 Nonlinear Classifiers via Basis Expansion

More information

CPSC 540: Machine Learning

CPSC 540: Machine Learning CPSC 540: Machine Learning Undirected Graphical Models Mark Schmidt University of British Columbia Winter 2016 Admin Assignment 3: 2 late days to hand it in today, Thursday is final day. Assignment 4:

More information

EXAM IN STATISTICAL MACHINE LEARNING STATISTISK MASKININLÄRNING

EXAM IN STATISTICAL MACHINE LEARNING STATISTISK MASKININLÄRNING EXAM IN STATISTICAL MACHINE LEARNING STATISTISK MASKININLÄRNING DATE AND TIME: June 9, 2018, 09.00 14.00 RESPONSIBLE TEACHER: Andreas Svensson NUMBER OF PROBLEMS: 5 AIDING MATERIAL: Calculator, mathematical

More information

SVAN 2016 Mini Course: Stochastic Convex Optimization Methods in Machine Learning

SVAN 2016 Mini Course: Stochastic Convex Optimization Methods in Machine Learning SVAN 2016 Mini Course: Stochastic Convex Optimization Methods in Machine Learning Mark Schmidt University of British Columbia, May 2016 www.cs.ubc.ca/~schmidtm/svan16 Some images from this lecture are

More information

Logistic Regression: Online, Lazy, Kernelized, Sequential, etc.

Logistic Regression: Online, Lazy, Kernelized, Sequential, etc. Logistic Regression: Online, Lazy, Kernelized, Sequential, etc. Harsha Veeramachaneni Thomson Reuter Research and Development April 1, 2010 Harsha Veeramachaneni (TR R&D) Logistic Regression April 1, 2010

More information

Gaussian Processes (10/16/13)

Gaussian Processes (10/16/13) STA561: Probabilistic machine learning Gaussian Processes (10/16/13) Lecturer: Barbara Engelhardt Scribes: Changwei Hu, Di Jin, Mengdi Wang 1 Introduction In supervised learning, we observe some inputs

More information

Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials

Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials Philipp Krähenbühl and Vladlen Koltun Stanford University Presenter: Yuan-Ting Hu 1 Conditional Random Field (CRF) E x I = φ u

More information

CSC321 Lecture 18: Learning Probabilistic Models

CSC321 Lecture 18: Learning Probabilistic Models CSC321 Lecture 18: Learning Probabilistic Models Roger Grosse Roger Grosse CSC321 Lecture 18: Learning Probabilistic Models 1 / 25 Overview So far in this course: mainly supervised learning Language modeling

More information

Regression. Goal: Learn a mapping from observations (features) to continuous labels given a training set (supervised learning)

Regression. Goal: Learn a mapping from observations (features) to continuous labels given a training set (supervised learning) Linear Regression Regression Goal: Learn a mapping from observations (features) to continuous labels given a training set (supervised learning) Example: Height, Gender, Weight Shoe Size Audio features

More information

The Representor Theorem, Kernels, and Hilbert Spaces

The Representor Theorem, Kernels, and Hilbert Spaces The Representor Theorem, Kernels, and Hilbert Spaces We will now work with infinite dimensional feature vectors and parameter vectors. The space l is defined to be the set of sequences f 1, f, f 3,...

More information

Iterative Laplacian Score for Feature Selection

Iterative Laplacian Score for Feature Selection Iterative Laplacian Score for Feature Selection Linling Zhu, Linsong Miao, and Daoqiang Zhang College of Computer Science and echnology, Nanjing University of Aeronautics and Astronautics, Nanjing 2006,

More information

Kernel Methods and Support Vector Machines

Kernel Methods and Support Vector Machines Kernel Methods and Support Vector Machines Oliver Schulte - CMPT 726 Bishop PRML Ch. 6 Support Vector Machines Defining Characteristics Like logistic regression, good for continuous input features, discrete

More information

COMS 4771 Lecture Boosting 1 / 16

COMS 4771 Lecture Boosting 1 / 16 COMS 4771 Lecture 12 1. Boosting 1 / 16 Boosting What is boosting? Boosting: Using a learning algorithm that provides rough rules-of-thumb to construct a very accurate predictor. 3 / 16 What is boosting?

More information