Machine Learning with Quantum-Inspired Tensor Networks
|
|
- Julianna Matthews
- 6 years ago
- Views:
Transcription
1 Machine Learning with Quantum-Inspired Tensor Networks E.M. Stoudenmire and David J. Schwab Advances in Neural Information Processing 29 arxiv: RIKEN AICS - Mar 2017
2 Collaboration with David J. Schwab, Northwestern and CUNY Graduate Center Quantum Machine Learning, Perimeter Institute, Aug 2016
3 Exciting time for machine learning Language Processing Self-driving cars Medicine Materials Science / Chemistry
4 Progress in neural networks and deep learning neural network diagram
5 Convolutional neural network "MERA" tensor network
6 Are tensor networks useful for machine learning? This Talk Tensor networks fit naturally into kernel learning (Also very strong connections to graphical models) Many benefits for learning Linear scaling Adaptive Feature sharing
7 Machine Learning Physics Topological Phases Phase Transitions Neural Nets Boltzmann Machines Quantum Monte Carlo Sign Problem Kernel Learning Materials Science & Chemistry Supervised Learning Unsupervised Learning Tensor Networks
8 Machine Learning Physics Topological Phases Phase Transitions Neural Nets Boltzmann Machines Quantum Monte Carlo Sign Problem Kernel Learning Materials Science & Chemistry Supervised Learning Unsupervised Learning (this talk) Tensor Networks
9 What are Tensor Networks?
10 How do tensor networks arise in physics? Quantum systems governed by Schrödinger equation: Ĥ ~ = E ~ It is just an eigenvalue problem.
11 The problem is that Ĥ is a 2 N x 2 N matrix ~ =) wavefunction has 2 N components = E Ĥ ~ ~
12 Natural to view wavefunction as order-n tensor i = X {s} s 1 s 2 s 3 s N s 1 s 2 s 3 s N i
13 Natural to view wavefunction as order-n tensor s 1 s 2 s 3 s 4 s N s 1 s 2 s 3 s N =
14 Tensor components related to probabilities of e.g. Ising model spin configurations ""#""#" = # # " " " " "
15 Tensor components related to probabilities of e.g. Ising model spin configurations # # # # # " " #""##"# =
16 Must find an approximation to this exponential problem s 1 s 2 s 3 s 4 s N s 1 s 2 s 3 s N =
17 Simplest approximation (mean field / rank-1) Let spins "do their own thing" s 1 s 2 s 3 s 4 s 5 s 6 ' s 1 s 2 s 3 s 4 s 5 s 6 s 1 s 2 s 3 s 4 s 5 s 6 Expected values of individual spins ok No correlations
18 Restore correlations locally s 1 s 2 s 3 s 4 s 5 s 6 ' s 1 s 2 s 3 s 4 s 5 s 6 s 1 s 2 s 3 s 4 s 5 s 6
19 Restore correlations locally s 1 s 2 s 3 s 4 s 5 s 6 ' s 1 s 2 s 3 s 4 s 5 s 6 i 1 i 1 s 1 s 2 s 3 s 4 s 5 s 6
20 Restore correlations locally s 1 s 2 s 3 s 4 s 5 s 6 ' s 1 s 2 s 3 s 4 s 5 s 6 i 1 i 1 i 2 i 2 i 3 i 3 i 4 i 4 i 5 i 5 s 1 s 2 s 3 s 4 s 5 s 6 matrix product state (MPS) Local expected values accurate Correlations decay with spatial distance
21 " # # " " # "Matrix product state" because retrieving an element = product of matrices
22 = "##""# "Matrix product state" because retrieving an element = product of matrices
23 Tensor diagrams have rigorous meaning j v j i j M ij i j k T ijk
24 Joining lines implies contraction, can omit names X i j j M ij v j A ij B jk = AB A ij B ji =Tr[AB]
25 MPS = matrix product state MPS approximation controlled by bond dimension "m" (like SVD rank) Compress 2 N N 2 m 2 parameters parameters into m 2 N 2 can represent any tensor
26 Friendly neighborhood of "quantum state space" m=8 m=4 m=2 m=1
27 MPS = matrix product state MPS lead to powerful optimization techniques (DMRG algorithm) White, PRL 69, 2863 (1992) Stoudenmire, White, PRB 87, (2013)
28 Besides MPS, other successful tensor are PEPS and MERA PEPS (2D systems) MERA (critical systems) Evenbly, Vidal, PRB 79, (2009) Verstraete, Cirac, cond-mat/ (2004) Orus, Ann. Phys. 349, 117 (2014)
29 Supervised Kernel Learning
30 Supervised Learning Very common task: Labeled training data (= supervised) Find decision function f(x) f(x) > 0 x 2 A f(x) < 0 x 2 B Input vector x e.g. image pixels
31 ML Overview Use training data to build model x 16 x 13 x 1 x 3 x 4 x 7 x 8 x 9 x 10 x 11 x 12 x 14 x 15 x 6 x 2 x 5
32 ML Overview Use training data to build model x 16 x 13 x 1 x 3 x 4 x 7 x 8 x 9 x 10 x 11 x 12 x 14 x 15 x 6 x 2 x 5
33 ML Overview Use training data to build model Generalize to unseen test data
34 ML Overview Popular approaches Neural Networks f(x) = 2 M 2 1 M 1 x Non-Linear Kernel Learning f(x) =W (x)
35 Non-linear kernel learning Want f(x) to separate classes Linear classifier often insufficient f(x) =W x??
36 Non-linear kernel learning Apply non-linear "feature map" x! (x)
37 Non-linear kernel learning Apply non-linear "feature map" x! (x) Decision function f(x) =W (x)
38 Non-linear kernel learning Decision function f(x) =W (x) Linear classifier in feature space
39 Non-linear kernel learning Example of feature map x =(x 1,x 2,x 3 ) (x) =(1,x 1,x 2,x 3,x 1 x 2,x 1 x 3,x 2 x 3 ) x is "lifted" to feature space
40 Proposal for Learning
41 Grayscale image data
42 Map pixels to "spins"
43 Map pixels to "spins"
44 Map pixels to "spins"
45 x = input Local feature map, dimension d=2 (x j )= h cos 2 x j i, sin 2 x j x j 2 [0, 1] Crucially, grayscale values not orthogonal
46 x = input = local feature map Total feature map (x) s 1 s 2 s N (x) = s 1 (x 1 ) s 2 (x 2 ) s N (x N ) Tensor product of local feature maps / vectors Just like product state wavefunction of spins Vector in dimensional space 2 N
47 x = input = local feature map Total feature map (x) More detailed notation x =[x 1, x 2, x 3,..., x N ] raw inputs (x) = [ 1( x 1 ) 2( x 1 ) [ [ 1( x 2 ) 2( x 2 ) [ [ 1( x 3 ) 2( x 3 ) [ [ 1( x N )... feature 2( x N ) [ vector
48 x = input = local feature map Total feature map (x) Tensor diagram notation x =[x 1, x 2, x 3,..., x N ] raw inputs (x) = s 1 s 2 s 3 s 4 s 5 s 6 s 1 s 2 s 3 s 4 s 5 s 6 s N s N feature vector
49 Construct decision function f(x) =W (x) (x)
50 Construct decision function f(x) =W (x) W (x)
51 Construct decision function f(x) =W (x) f(x) = W (x)
52 Construct decision function f(x) =W (x) f(x) = W (x) W =
53 Main approximation W = order-n tensor matrix product state (MPS)
54 MPS form of decision function f(x) = W (x)
55 Linear scaling Can use algorithm similar to DMRG to optimize Scaling is N N T m 3 N = size of input N T = size of training set m = MPS bond dimension f(x) = W (x)
56 Linear scaling Can use algorithm similar to DMRG to optimize Scaling is N N T m 3 N = size of input N T = size of training set m = MPS bond dimension f(x) = W (x)
57 Linear scaling Can use algorithm similar to DMRG to optimize Scaling is N N T m 3 N = size of input N T = size of training set m = MPS bond dimension f(x) = W (x)
58 Linear scaling Can use algorithm similar to DMRG to optimize Scaling is N N T m 3 N = size of input N T = size of training set m = MPS bond dimension f(x) = W (x)
59 Linear scaling Can use algorithm similar to DMRG to optimize Scaling is N N T m 3 N = size of input N T = size of training set m = MPS bond dimension f(x) = W (x) Could improve with stochastic gradient
60 Multi-class extension of model Decision function f `(x) =W ` (x) Index ` runs over possible labels ` f `(x) = W ` (x) ` = W ` (x) Predicted label is argmax` f `(x)
61 MNIST Experiment MNIST is a benchmark data set of grayscale handwritten digits (labels ` = 0,1,2,...,9) 60,000 labeled training images 10,000 labeled test images
62 MNIST Experiment One-dimensional mapping
63 MNIST Experiment Results Bond dimension m = 10 m = 20 m = 120 Test Set Error ~5% (500/10,000 incorrect) ~2% (200/10,000 incorrect) 0.97% (97/10,000 incorrect) State of the art is < 1% test set error
64 MNIST Experiment Demo Link:
65 Understanding Tensor Network Models f(x) = W (x)
66 Again assume W is an MPS f(x) = W (x) Many interesting benefits Two are: 1. Adaptive 2. Feature sharing
67 { 1. Tensor networks are adaptive boundary pixels not useful for learning grayscale training data
68 2. Feature sharing ` f `(x) = W ` (x) ` = Different central tensors "Wings" shared between models Regularizes models
69 2. Feature sharing ` f `(x) = Progressively learn shared features
70 2. Feature sharing ` f `(x) = Progressively learn shared features
71 2. Feature sharing ` f `(x) = Progressively learn shared features
72 2. Feature sharing ` f `(x) = Progressively learn shared features Deliver to central tensor `
73 Nature of Weight Tensor Representer theorem says exact W = X j j (x j ) Density plots of trained W ` for each label ` =0, 1,...,9
74 Nature of Weight Tensor Representer theorem says exact W = X j j (x j ) Tensor network approx. can violate this condition W MPS 6= X j j (x j ) for any { j } Tensor network learning not interpolation Interesting consequences for generalization?
75 Some Future Directions Apply to 1D data sets (audio, time series) Other tensor networks: TTN, PEPS, MERA Useful to interpret W (x) 2 as probability? Could import even more physics insights. Features extracted by elements of tensor network?
76 W What functions realized for arbitrary? Instead of "spin" local feature map, use* (x) =(1,x) Recall total feature map is [ (x) = 1 ( x 1 ) 2( x 1 ) [ [ 1( x 2 ) 2( x 2 ) [ [ 1( x 3 ) 2( x 3 ) [... [ 1( x N ) 2( x N ) [ *Novikov, et al., arxiv:
77 [ N=2 case (x) =(1,x) (x) =[ 1 [ x 1 [ 1 x 2 =(1,x 1,x 2,x 1 x 2 ) (W 11,W 21,W 12,W 22 ) f(x) =W (x) = (1, x 1, x 2, x 1 x 2 ) = W 11 + W 21 x 1 + W 12 x 2 + W 22 x 1 x 2
78 [ [ N=3 case (x) =(1,x) (x) =[ 1 [ x 1 [ 1 x 2 [ 1 x 3 =(1,x 1,x 2,x 3,x 1 x 2,x 1 x 3,x 2 x 3,x 1 x 2 x 3 ) f(x) =W (x) = W W 211 x 1 + W 121 x 2 + W 112 x 3 + W 221 x 1 x 2 + W 212 x 1 x 3 + W 122 x 1 x 3 + W 222 x 1 x 2 x 3
79 General N case x 2 R N f(x) =W (x) = W W x 1 + W x 2 + W x W x 1 x 2 + W x 1 x W x 1 x 2 x constant singles doubles triples W x 1 x 2 x 3 x N N-tuple Model has exponentially many formal parameters Novikov, Trofimov, Oseledets, arxiv: (2016)
80 Related Work Novikov, Trofimov, Oseledets ( ) matrix product states + kernel learning stochastic gradient descent Cohen, Sharir, Shashua ( , , , ) tree tensor networks expressivity of tensor network models correlations of data (analogue of entanglement entropy) generative proposal
81 Other MPS related work ( = "tensor trains") Markov random field models Novikov et al., Proceedings of 31st ICML (2014) Large scale PCA Lee, Cichocki, arxiv: (2014) Feature extraction of tensor data Bengua et al., IEEE Congress on Big Data (2015) Compressing weights of neural nets Novikov et al., Advances in Neural Information Processing (2015)
Machine Learning with Tensor Networks
Machine Learning with Tensor Networks E.M. Stoudenmire and David J. Schwab Advances in Neural Information Processing 29 arxiv:1605.05775 Beijing Jun 2017 Machine learning has physics in its DNA # " # #
More informationSupervised Learning with Tensor Networks
Supervised Learning with Tensor Networks E. M. Stoudenmire Perimeter Institute for Theoretical Physics Waterloo, Ontario, N2L 2Y5, Canada David J. Schwab Department of Physics Northwestern University,
More informationTensor network vs Machine learning. Song Cheng ( 程嵩 ) IOP, CAS
Tensor network vs Machine learning Song Cheng ( 程嵩 ) IOP, CAS physichengsong@iphy.ac.cn Outline Tensor network in a nutshell TN concepts in machine learning TN methods in machine learning Outline Tensor
More informationIntroduction to tensor network state -- concept and algorithm. Z. Y. Xie ( 谢志远 ) ITP, Beijing
Introduction to tensor network state -- concept and algorithm Z. Y. Xie ( 谢志远 ) 2018.10.29 ITP, Beijing Outline Illusion of complexity of Hilbert space Matrix product state (MPS) as lowly-entangled state
More informationThe density matrix renormalization group and tensor network methods
The density matrix renormalization group and tensor network methods Outline Steve White Exploiting the low entanglement of ground states Matrix product states and DMRG 1D 2D Tensor network states Some
More informationQuantum simulation with string-bond states: Joining PEPS and Monte Carlo
Quantum simulation with string-bond states: Joining PEPS and Monte Carlo N. Schuch 1, A. Sfondrini 1,2, F. Mezzacapo 1, J. Cerrillo 1,3, M. Wolf 1,4, F. Verstraete 5, I. Cirac 1 1 Max-Planck-Institute
More informationQuantum Convolutional Neural Networks
Quantum Convolutional Neural Networks Iris Cong Soonwon Choi Mikhail D. Lukin arxiv:1810.03787 Berkeley Quantum Information Seminar October 16 th, 2018 Why quantum machine learning? Machine learning: interpret
More informationCS 6375 Machine Learning
CS 6375 Machine Learning Nicholas Ruozzi University of Texas at Dallas Slides adapted from David Sontag and Vibhav Gogate Course Info. Instructor: Nicholas Ruozzi Office: ECSS 3.409 Office hours: Tues.
More informationTensor network methods in condensed matter physics. ISSP, University of Tokyo, Tsuyoshi Okubo
Tensor network methods in condensed matter physics ISSP, University of Tokyo, Tsuyoshi Okubo Contents Possible target of tensor network methods! Tensor network methods! Tensor network states as ground
More informationIntroduction to the Tensor Train Decomposition and Its Applications in Machine Learning
Introduction to the Tensor Train Decomposition and Its Applications in Machine Learning Anton Rodomanov Higher School of Economics, Russia Bayesian methods research group (http://bayesgroup.ru) 14 March
More informationAn introduction to tensornetwork
An introduction to tensornetwork states and MERA Sissa Journal Club Andrea De Luca 29/01/2010 A typical problem We are given: A lattice with N sites On each site a C d hilbert space A quantum hamiltonian
More information4 Matrix product states
Physics 3b Lecture 5 Caltech, 05//7 4 Matrix product states Matrix product state (MPS) is a highly useful tool in the study of interacting quantum systems in one dimension, both analytically and numerically.
More information3 Symmetry Protected Topological Phase
Physics 3b Lecture 16 Caltech, 05/30/18 3 Symmetry Protected Topological Phase 3.1 Breakdown of noninteracting SPT phases with interaction Building on our previous discussion of the Majorana chain and
More informationLoop optimization for tensor network renormalization
Yukawa Institute for Theoretical Physics, Kyoto University, Japan June, 6 Loop optimization for tensor network renormalization Shuo Yang! Perimeter Institute for Theoretical Physics, Waterloo, Canada Zheng-Cheng
More informationIntroduction to Tensor Networks: PEPS, Fermions, and More
Introduction to Tensor Networks: PEPS, Fermions, and More Román Orús Institut für Physik, Johannes Gutenberg-Universität, Mainz (Germany)! School on computational methods in quantum materials Jouvence,
More informationFermionic tensor networks
Fermionic tensor networks Philippe Corboz, ETH Zurich / EPF Lausanne, Switzerland Collaborators: University of Queensland: Guifre Vidal, Roman Orus, Glen Evenbly, Jacob Jordan University of Vienna: Frank
More informationarxiv: v2 [cs.lg] 10 Apr 2017
Deep Learning and Quantum Entanglement: Fundamental Connections with Implications to Network Design arxiv:1704.01552v2 [cs.lg] 10 Apr 2017 Yoav Levine David Yakira Nadav Cohen Amnon Shashua The Hebrew
More informationClassification with Perceptrons. Reading:
Classification with Perceptrons Reading: Chapters 1-3 of Michael Nielsen's online book on neural networks covers the basics of perceptrons and multilayer neural networks We will cover material in Chapters
More informationIPAM/UCLA, Sat 24 th Jan Numerical Approaches to Quantum Many-Body Systems. QS2009 tutorials. lecture: Tensor Networks.
IPAM/UCLA, Sat 24 th Jan 2009 umerical Approaches to Quantum Many-Body Systems QS2009 tutorials lecture: Tensor etworks Guifre Vidal Outline Tensor etworks Computation of expected values Optimization of
More informationJournal Club: Brief Introduction to Tensor Network
Journal Club: Brief Introduction to Tensor Network Wei-Han Hsiao a a The University of Chicago E-mail: weihanhsiao@uchicago.edu Abstract: This note summarizes the talk given on March 8th 2016 which was
More informationT ensor N et works. I ztok Pizorn Frank Verstraete. University of Vienna M ichigan Quantum Summer School
T ensor N et works I ztok Pizorn Frank Verstraete University of Vienna 2010 M ichigan Quantum Summer School Matrix product states (MPS) Introduction to matrix product states Ground states of finite systems
More informationFermionic tensor networks
Fermionic tensor networks Philippe Corboz, Institute for Theoretical Physics, ETH Zurich Bosons vs Fermions P. Corboz and G. Vidal, Phys. Rev. B 80, 165129 (2009) : fermionic 2D MERA P. Corboz, R. Orus,
More informationNeural Network Representation of Tensor Network and Chiral States
Neural Network Representation of Tensor Network and Chiral States Yichen Huang ( 黄溢辰 ) 1 and Joel E. Moore 2 1 Institute for Quantum Information and Matter California Institute of Technology 2 Department
More informationThe Perceptron. Volker Tresp Summer 2016
The Perceptron Volker Tresp Summer 2016 1 Elements in Learning Tasks Collection, cleaning and preprocessing of training data Definition of a class of learning models. Often defined by the free model parameters
More informationSimulation of Quantum Many-Body Systems
Numerical Quantum Simulation of Matteo Rizzi - KOMET 7 - JGU Mainz Vorstellung der Arbeitsgruppen WS 15-16 recent developments in control of quantum objects (e.g., cold atoms, trapped ions) General Framework
More informationUnderstanding Deep Learning via Physics:
Understanding Deep Learning via Physics: The use of Quantum Entanglement for studying the Inductive Bias of Convolutional Networks Nadav Cohen Institute for Advanced Study Symposium on Physics and Machine
More informationLecture 3: Tensor Product Ansatz
Lecture 3: Tensor Product nsatz Graduate Lectures Dr Gunnar Möller Cavendish Laboratory, University of Cambridge slide credits: Philippe Corboz (ETH / msterdam) January 2014 Cavendish Laboratory Part I:
More informationOpportunities and challenges in quantum-enhanced machine learning in near-term quantum computers
Opportunities and challenges in quantum-enhanced machine learning in near-term quantum computers Alejandro Perdomo-Ortiz Senior Research Scientist, Quantum AI Lab. at NASA Ames Research Center and at the
More informationMatrix-Product states: Properties and Extensions
New Development of Numerical Simulations in Low-Dimensional Quantum Systems: From Density Matrix Renormalization Group to Tensor Network Formulations October 27-29, 2010, Yukawa Institute for Theoretical
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 informationCS 179: LECTURE 16 MODEL COMPLEXITY, REGULARIZATION, AND CONVOLUTIONAL NETS
CS 179: LECTURE 16 MODEL COMPLEXITY, REGULARIZATION, AND CONVOLUTIONAL NETS LAST TIME Intro to cudnn Deep neural nets using cublas and cudnn TODAY Building a better model for image classification Overfitting
More informationNeural Networks. Single-layer neural network. CSE 446: Machine Learning Emily Fox University of Washington March 10, /9/17
3/9/7 Neural Networks Emily Fox University of Washington March 0, 207 Slides adapted from Ali Farhadi (via Carlos Guestrin and Luke Zettlemoyer) Single-layer neural network 3/9/7 Perceptron as a neural
More informationAdvanced Computation for Complex Materials
Advanced Computation for Complex Materials Computational Progress is brainpower limited, not machine limited Algorithms Physics Major progress in algorithms Quantum Monte Carlo Density Matrix Renormalization
More informationRenormalization of Tensor Network States
Renormalization of Tensor Network States I. Coarse Graining Tensor Renormalization Tao Xiang Institute of Physics Chinese Academy of Sciences txiang@iphy.ac.cn Numerical Renormalization Group brief introduction
More informationMachine Learning Techniques
Machine Learning Techniques ( 機器學習技法 ) Lecture 13: Deep Learning Hsuan-Tien Lin ( 林軒田 ) htlin@csie.ntu.edu.tw Department of Computer Science & Information Engineering National Taiwan University ( 國立台灣大學資訊工程系
More informationExcursion: MPS & DMRG
Excursion: MPS & DMRG Johannes.Schachenmayer@gmail.com Acronyms for: - Matrix product states - Density matrix renormalization group Numerical methods for simulations of time dynamics of large 1D quantum
More informationQuantum many-body systems and tensor networks: simulation methods and applications
Quantum many-body systems and tensor networks: simulation methods and applications Román Orús School of Physical Sciences, University of Queensland, Brisbane (Australia) Department of Physics and Astronomy,
More informationTensor network simulation of QED on infinite lattices: learning from (1 + 1)d, and prospects for (2 + 1)d
Tensor network simulation of QED on infinite lattices: learning from (1 + 1)d, and prospects for (2 + 1)d Román Orús University of Mainz (Germany) K. Zapp, RO, Phys. Rev. D 95, 114508 (2017) Goal of this
More informationConvolution and Pooling as an Infinitely Strong Prior
Convolution and Pooling as an Infinitely Strong Prior Sargur Srihari srihari@buffalo.edu This is part of lecture slides on Deep Learning: http://www.cedar.buffalo.edu/~srihari/cse676 1 Topics in Convolutional
More informationMatrix Product States
Matrix Product States Ian McCulloch University of Queensland Centre for Engineered Quantum Systems 28 August 2017 Hilbert space (Hilbert) space is big. Really big. You just won t believe how vastly, hugely,
More informationCSC321 Lecture 20: Autoencoders
CSC321 Lecture 20: Autoencoders Roger Grosse Roger Grosse CSC321 Lecture 20: Autoencoders 1 / 16 Overview Latent variable models so far: mixture models Boltzmann machines Both of these involve discrete
More informationLearning From Data Lecture 15 Reflecting on Our Path - Epilogue to Part I
Learning From Data Lecture 15 Reflecting on Our Path - Epilogue to Part I What We Did The Machine Learning Zoo Moving Forward M Magdon-Ismail CSCI 4100/6100 recap: Three Learning Principles Scientist 2
More informationSimon Trebst. Quantitative Modeling of Complex Systems
Quantitative Modeling of Complex Systems Machine learning In computer science, machine learning is concerned with algorithms that allow for data analytics, most prominently dimensional reduction and feature
More informationDeep learning / Ian Goodfellow, Yoshua Bengio and Aaron Courville. - Cambridge, MA ; London, Spis treści
Deep learning / Ian Goodfellow, Yoshua Bengio and Aaron Courville. - Cambridge, MA ; London, 2017 Spis treści Website Acknowledgments Notation xiii xv xix 1 Introduction 1 1.1 Who Should Read This Book?
More informationThe Perceptron. Volker Tresp Summer 2018
The Perceptron Volker Tresp Summer 2018 1 Elements in Learning Tasks Collection, cleaning and preprocessing of training data Definition of a class of learning models. Often defined by the free model parameters
More informationTensor networks and deep learning
Tensor networks and deep learning I. Oseledets, A. Cichocki Skoltech, Moscow 26 July 2017 What is a tensor Tensor is d-dimensional array: A(i 1,, i d ) Why tensors Many objects in machine learning can
More informationFermionic topological quantum states as tensor networks
order in topological quantum states as Jens Eisert, Freie Universität Berlin Joint work with Carolin Wille and Oliver Buerschaper Symmetry, topology, and quantum phases of matter: From to physical realizations,
More informationTypical quantum states at finite temperature
Typical quantum states at finite temperature How should one think about typical quantum states at finite temperature? Density Matrices versus pure states Why eigenstates are not typical Measuring the heat
More informationTime Evolving Block Decimation Algorithm
Time Evolving Block Decimation Algorithm Application to bosons on a lattice Jakub Zakrzewski Marian Smoluchowski Institute of Physics and Mark Kac Complex Systems Research Center, Jagiellonian University,
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 informationExpressive Efficiency and Inductive Bias of Convolutional Networks:
Expressive Efficiency and Inductive Bias of Convolutional Networks: Analysis & Design via Hierarchical Tensor Decompositions Nadav Cohen The Hebrew University of Jerusalem AAAI Spring Symposium Series
More informationNon-Convex Optimization in Machine Learning. Jan Mrkos AIC
Non-Convex Optimization in Machine Learning Jan Mrkos AIC The Plan 1. Introduction 2. Non convexity 3. (Some) optimization approaches 4. Speed and stuff? Neural net universal approximation Theorem (1989):
More informationExpressive Efficiency and Inductive Bias of Convolutional Networks:
Expressive Efficiency and Inductive Bias of Convolutional Networks: Analysis & Design via Hierarchical Tensor Decompositions Nadav Cohen The Hebrew University of Jerusalem AAAI Spring Symposium Series
More informationMachine learning quantum phases of matter
Machine learning quantum phases of matter Summer School on Emergent Phenomena in Quantum Materials Cornell, May 2017 Simon Trebst University of Cologne Machine learning quantum phases of matter Summer
More informationSum-Product Networks: A New Deep Architecture
Sum-Product Networks: A New Deep Architecture Pedro Domingos Dept. Computer Science & Eng. University of Washington Joint work with Hoifung Poon 1 Graphical Models: Challenges Bayesian Network Markov Network
More informationHow to do backpropagation in a brain
How to do backpropagation in a brain Geoffrey Hinton Canadian Institute for Advanced Research & University of Toronto & Google Inc. Prelude I will start with three slides explaining a popular type of deep
More informationComputer 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 informationQuantum Hamiltonian Complexity. Itai Arad
1 18 / Quantum Hamiltonian Complexity Itai Arad Centre of Quantum Technologies National University of Singapore QIP 2015 2 18 / Quantum Hamiltonian Complexity condensed matter physics QHC complexity theory
More informationHolographic Geometries from Tensor Network States
Holographic Geometries from Tensor Network States J. Molina-Vilaplana 1 1 Universidad Politécnica de Cartagena Perspectives on Quantum Many-Body Entanglement, Mainz, Sep 2013 1 Introduction & Motivation
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 informationECE 521. Lecture 11 (not on midterm material) 13 February K-means clustering, Dimensionality reduction
ECE 521 Lecture 11 (not on midterm material) 13 February 2017 K-means clustering, Dimensionality reduction With thanks to Ruslan Salakhutdinov for an earlier version of the slides Overview K-means clustering
More informationTensor network simulations of strongly correlated quantum systems
CENTRE FOR QUANTUM TECHNOLOGIES NATIONAL UNIVERSITY OF SINGAPORE AND CLARENDON LABORATORY UNIVERSITY OF OXFORD Tensor network simulations of strongly correlated quantum systems Stephen Clark LXXT[[[GSQPEFS\EGYOEGXMZMXMIWUYERXYQGSYVWI
More informationPositive Tensor Network approach for simulating open quantum many-body systems
Positive Tensor Network approach for simulating open quantum many-body systems 19 / 9 / 2016 A. Werner, D. Jaschke, P. Silvi, M. Kliesch, T. Calarco, J. Eisert and S. Montangero PRL 116, 237201 (2016)
More informationGuiding Monte Carlo Simulations with Machine Learning
Guiding Monte Carlo Simulations with Machine Learning Yang Qi Department of Physics, Massachusetts Institute of Technology Joining Fudan University in 2017 KITS, June 29 2017. 1/ 33 References Works led
More informationSupervised Learning. George Konidaris
Supervised Learning George Konidaris gdk@cs.brown.edu Fall 2017 Machine Learning Subfield of AI concerned with learning from data. Broadly, using: Experience To Improve Performance On Some Task (Tom Mitchell,
More informationMachine 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 informationStatistical Machine Learning Theory. From Multi-class Classification to Structured Output Prediction. Hisashi Kashima.
http://goo.gl/xilnmn Course website KYOTO UNIVERSITY Statistical Machine Learning Theory From Multi-class Classification to Structured Output Prediction Hisashi Kashima kashima@i.kyoto-u.ac.jp DEPARTMENT
More informationIntroduction to Machine Learning
Introduction to Machine Learning Neural Networks Varun Chandola x x 5 Input Outline Contents February 2, 207 Extending Perceptrons 2 Multi Layered Perceptrons 2 2. Generalizing to Multiple Labels.................
More informationSimulation of Quantum Many-Body Systems
Numerical Quantum Simulation of Matteo Rizzi - KOMET 337 - JGU Mainz Vorstellung der Arbeitsgruppen WS 14-15 QMBS: An interdisciplinary topic entanglement structure of relevant states anyons for q-memory
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 informationThe Perceptron. Volker Tresp Summer 2014
The Perceptron Volker Tresp Summer 2014 1 Introduction One of the first serious learning machines Most important elements in learning tasks Collection and preprocessing of training data Definition of a
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 informationTopic 3: Neural Networks
CS 4850/6850: Introduction to Machine Learning Fall 2018 Topic 3: Neural Networks Instructor: Daniel L. Pimentel-Alarcón c Copyright 2018 3.1 Introduction Neural networks are arguably the main reason why
More informationMarkov Chains for Tensor Network States
Sofyan Iblisdir University of Barcelona September 2013 arxiv 1309.4880 Outline Main issue discussed here: how to find good tensor network states? The optimal state problem Markov chains Multiplicative
More informationLecture 16 Deep Neural Generative Models
Lecture 16 Deep Neural Generative Models CMSC 35246: Deep Learning Shubhendu Trivedi & Risi Kondor University of Chicago May 22, 2017 Approach so far: We have considered simple models and then constructed
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 informationRenormalization of Tensor- Network States Tao Xiang
Renormalization of Tensor- Network States Tao Xiang Institute of Physics/Institute of Theoretical Physics Chinese Academy of Sciences txiang@iphy.ac.cn Physical Background: characteristic energy scales
More informationThe Origin of Deep Learning. Lili Mou Jan, 2015
The Origin of Deep Learning Lili Mou Jan, 2015 Acknowledgment Most of the materials come from G. E. Hinton s online course. Outline Introduction Preliminary Boltzmann Machines and RBMs Deep Belief Nets
More informationMatrix Product Operators: Algebras and Applications
Matrix Product Operators: Algebras and Applications Frank Verstraete Ghent University and University of Vienna Nick Bultinck, Jutho Haegeman, Michael Marien Burak Sahinoglu, Dominic Williamson Ignacio
More informationLearning to Disentangle Factors of Variation with Manifold Learning
Learning to Disentangle Factors of Variation with Manifold Learning Scott Reed Kihyuk Sohn Yuting Zhang Honglak Lee University of Michigan, Department of Electrical Engineering and Computer Science 08
More informationAlgorithms other than SGD. CS6787 Lecture 10 Fall 2017
Algorithms other than SGD CS6787 Lecture 10 Fall 2017 Machine learning is not just SGD Once a model is trained, we need to use it to classify new examples This inference task is not computed with SGD There
More informationAdvanced computational methods X Selected Topics: SGD
Advanced computational methods X071521-Selected Topics: SGD. In this lecture, we look at the stochastic gradient descent (SGD) method 1 An illustrating example The MNIST is a simple dataset of variety
More informationStatistical Machine Learning
Statistical Machine Learning Lecture 9 Numerical optimization and deep learning Niklas Wahlström Division of Systems and Control Department of Information Technology Uppsala University niklas.wahlstrom@it.uu.se
More informationEntanglement spectrum and Matrix Product States
Entanglement spectrum and Matrix Product States Frank Verstraete J. Haegeman, D. Draxler, B. Pirvu, V. Stojevic, V. Zauner, I. Pizorn I. Cirac (MPQ), T. Osborne (Hannover), N. Schuch (Aachen) Outline Valence
More informationStatistical Machine Learning Theory. From Multi-class Classification to Structured Output Prediction. Hisashi Kashima.
http://goo.gl/jv7vj9 Course website KYOTO UNIVERSITY Statistical Machine Learning Theory From Multi-class Classification to Structured Output Prediction Hisashi Kashima kashima@i.kyoto-u.ac.jp DEPARTMENT
More informationMath 671: Tensor Train decomposition methods II
Math 671: Tensor Train decomposition methods II Eduardo Corona 1 1 University of Michigan at Ann Arbor December 13, 2016 Table of Contents 1 What we ve talked about so far: 2 The Tensor Train decomposition
More informationIntroduction to Machine Learning
Introduction to Machine Learning CS4731 Dr. Mihail Fall 2017 Slide content based on books by Bishop and Barber. https://www.microsoft.com/en-us/research/people/cmbishop/ http://web4.cs.ucl.ac.uk/staff/d.barber/pmwiki/pmwiki.php?n=brml.homepage
More informationSTA 4273H: Statistical Machine Learning
STA 4273H: Statistical Machine Learning Russ Salakhutdinov Department of Statistics! rsalakhu@utstat.toronto.edu! http://www.utstat.utoronto.ca/~rsalakhu/ Sidney Smith Hall, Room 6002 Lecture 3 Linear
More informationarxiv: v1 [cond-mat.str-el] 19 Jan 2012
Perfect Sampling with Unitary Tensor Networks arxiv:1201.3974v1 [cond-mat.str-el] 19 Jan 2012 Andrew J. Ferris 1, 2 and Guifre Vidal 1, 3 1 The University of Queensland, School of Mathematics and Physics,
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 informationSimulating Quantum Systems through Matrix Product States. Laura Foini SISSA Journal Club
Simulating Quantum Systems through Matrix Product States Laura Foini SISSA Journal Club 15-04-2010 Motivations Theoretical interest in Matrix Product States Wide spectrum of their numerical applications
More informationarxiv: v4 [quant-ph] 13 Mar 2013
Deep learning and the renormalization group arxiv:1301.3124v4 [quant-ph] 13 Mar 2013 Cédric Bény Institut für Theoretische Physik Leibniz Universität Hannover Appelstraße 2, 30167 Hannover, Germany cedric.beny@gmail.com
More informationIntroduction to Machine Learning. Introduction to ML - TAU 2016/7 1
Introduction to Machine Learning Introduction to ML - TAU 2016/7 1 Course Administration Lecturers: Amir Globerson (gamir@post.tau.ac.il) Yishay Mansour (Mansour@tau.ac.il) Teaching Assistance: Regev Schweiger
More informationIntroduction to Machine learning
Introduction to Machine learning Some slides and images are taken from: David Wolfe Corne Wikipedia Geoffrey A. Hinton https://www.macs.hw.ac.uk/~dwcorne/teaching/introdl.ppt Examples 1 Examples 1 WaveNet
More informationMachine Learning Basics
Security and Fairness of Deep Learning Machine Learning Basics Anupam Datta CMU Spring 2019 Image Classification Image Classification Image classification pipeline Input: A training set of N images, each
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 informationTUTORIAL PART 1 Unsupervised Learning
TUTORIAL PART 1 Unsupervised Learning Marc'Aurelio Ranzato Department of Computer Science Univ. of Toronto ranzato@cs.toronto.edu Co-organizers: Honglak Lee, Yoshua Bengio, Geoff Hinton, Yann LeCun, Andrew
More informationNumerical tensor methods and their applications
Numerical tensor methods and their applications 8 May 2013 All lectures 4 lectures, 2 May, 08:00-10:00: Introduction: ideas, matrix results, history. 7 May, 08:00-10:00: Novel tensor formats (TT, HT, QTT).
More informationHolographic Branching and Entanglement Renormalization
KITP, December 7 th 2010 Holographic Branching and Entanglement Renormalization Glen Evenbly Guifre Vidal Tensor Network Methods (DMRG, PEPS, TERG, MERA) Potentially offer general formalism to efficiently
More informationDeep Learning Autoencoder Models
Deep Learning Autoencoder Models Davide Bacciu Dipartimento di Informatica Università di Pisa Intelligent Systems for Pattern Recognition (ISPR) Generative Models Wrap-up Deep Learning Module Lecture Generative
More information