1 of 50

1

A Tensor is an 𝑑-way array

 

 

 

 

2 of 50

If you need more information, these references are helpful:�

  1. Kolda, T. G., & Bader, B. W. (2009). Tensor decompositions and applications. SIAM review, 51(3), 455-500.
  2. Matlab Tensor Toolbox: https://www.tensortoolbox.org/

The following slides contain material from Tammy Kolda, Charles Van Loan and Lars Elden

Tensor Decompositions

 

3 of 50

3

Grid Cells: Global positioning system that helps people navigate

4 of 50

4

5 of 50

5

Please Read!

6 of 50

6

Inside the Nucleus: Chromosome Territories

Tan, Longzhi, et al. "Three-dimensional genome structures of single diploid human cells." Science 361.6405 (2018): 924-928.

7 of 50

7

Some Mathematics….

8 of 50

8

The Khatri-Rao Product

Definition:

C.R. Rao and S.K. Mitra (1971). Generalized Inverse of Matrices and Applications, John Wiley and Sons, New York

9 of 50

9

The Khatri-Rao Product

If

then the Khatri-Rao product of B and C is given by

Note:

10 of 50

10

The Kronecker Product

11 of 50

11

The Kronecker Product

12 of 50

12

All Possible Entry- Entry Products

13 of 50

13

Kronecker Products of Kronecker Products

14 of 50

14

You have seen them before

15 of 50

15

Block matrix

16 of 50

16

Block Matrices: Addition

17 of 50

17

Block Matrix: Multiplication

18 of 50

18

Block Matrix: Transposition

19 of 50

19

A block matrix can have block matrix entries

20 of 50

20

Reshape

21 of 50

21

Reshape

22 of 50

22

SVD

23 of 50

23

SVD

24 of 50

24

The Tucker product

Definition:

25 of 50

25

The Tucker product representation

Before…

Now…

26 of 50

26

The Higher-Order SVD (HOSVD)

27 of 50

27

The HOSVD of a matrix

28 of 50

28

The HOSVD of a matrix

29 of 50

29

Notations

30 of 50

30

Tensor Unfolding

31 of 50

31

A Common Framework for Tensor Computations

Tensor Unfolding

32 of 50

32

Illustration of row-wise and column-wise unfolding (flattening, matricizing) of a third-order tensor

Andrzej Cichocki, Rafal Zdunek, and Shun'ichi Amari. Nonnegative Matrix and Tensor Factorizations: Applications to Exploratory Multi-way Data Analysis and Blind Source (2009)

33 of 50

33

Unfolding (matricizing) of a third-order tensor

34 of 50

34

SVD Expansion

35 of 50

Rank One Tensors

Building block for decomposition: Vector outer products = Rank one tensors

Matrix version (2 – way)

Rank one matrix

Tensor version (3 – way)

Rank one tensor

36 of 50

Building block for decomposition: Vector outer products = Rank one tensors

 

Visualization

gets weird.. Math is still good!

Rank one Tensors

37 of 50

37

Data

Low-rank model

Matrix notation

Sum of squared errors

Low Rank Approximation

38 of 50

Tensor Decompositions

CP Model: Sum of d-way outer products, useful for interpretation

Tucker Model: Project onto high-variance subspaces to reduce dimensionality

39 of 50

CP Decomposition

The full name is CANDECOMP/PARAFAC Decomposition

CANDECOMP = Canonical Decomposition

PARAFAC = Parallel Factors Decomposition

CP Tensor Factorization (3-way): Detecting low-rank 3-way structure

Higher-order analogue of the SVD

CP first invented in 1927

F. L. Hitchcock, The Expression of a Tensor or a Polyadic as a Sum of Products, Journal of Mathematics and Physics, 1927

40 of 50

Tensor notation

Sum of squared errors

Factor matrices

Component

CP Decomposition

41 of 50

Example 1: Neuronal Activity

Williams, A. H., Kim, T. H., Wang, F., Vyas, S., Ryu, S. I., Shenoy, K. V., ... & Ganguli, S. (2018). Unsupervised discovery of demixed, low-dimensional neural dynamics across multiple timescales through tensor component analysis. Neuron, 98(6), 1099-1115.

42 of 50

Example 1: Neuronal activity

Mouse

in “maze”

Neural activity

One Column of Neuron x

Time matrix

 

× 600 trials (over 5 days)

Microscope by

Inscopix

43 of 50

Trials vary start position and strategies

  • 600 Trials over 5 days
  • Start West or East
  • Conditions swap twice
    • Always turn South
    • Always turn Right
    • Always turn South

Example: Neuronal activity

44 of 50

8-Component CP decomposition of data

45 of 50

8-Component CP decomposition of data

46 of 50

Got the treat! Happy neurons….

8-Component CP decomposition of data

47 of 50

8-Component CP decomposition of data

Clear separation. Starting position Neurons understand mouse has started. ….

48 of 50

8-Component CP decomposition of data

49 of 50

Example 2: Application to hazardous gas

 

Vergara, A., Fonollosa, J., Mahiques, J., Trincavelli, M., Rulkov, N., & Huerta, R. (2013). On the performance of gas sensor arrays in open sampling systems using Inhibitory Support Vector Machines. Sensors and Actuators B: Chemical, 185, 462-477.

50 of 50

Factors

Example 2: Application to hazardous gas