SlideShare a Scribd company logo
1 of 28
Download to read offline
Attacking the Curse of Dimensionality
using Sums of Separable Functions
Martin J. Mohlenkamp
Department of Mathematics
http://www.ohiouniversityfaculty.com/mohlenka/
SAMSI, September 2019
Abstract
Naive computations involving a function of many variables suffer from the
curse of dimensionality: the computational cost grows exponentially with
the number of variables. One approach to bypassing the curse is to
approximate the function as a sum of products of functions of one variable
and compute in this format. When the variables are indices, a function of
many variables is called a tensor, and this approach is to approximate and
use the tensor in the (so-called) canonical tensor format. In this talk I will
describe how such approximations can be used in numerical analysis and in
machine learning.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 2 / 28
Goals of this Talk
Show you a tool that you may find useful.
Hint at other things I know that you may find useful.
Not Goals
Convince you that this tool is better than other methods.
Show that I am great.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 3 / 28
The Curse of Dimensionality (discrete setting)
d Name Notation Storage Visual
1 Vector vj $
2 Matrix Ajk $$
3 Tensor Tjkm $$$
> 3 Tensor T(j1, . . . , jd ) $d ?
The cost to do anything, even store the object,
grows exponentially in the dimension d.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 4 / 28
The Curse of Dimensionality (function setting)
To approximate a function f (x1, x2, . . . , xd )
that has smoothness p
to accuracy
costs −d/p = ( −1/p)d = $d .
This curse is unavoidable for general function spaces (smoothness classes).
If a method seems to avoid it, look for
“constants” that grow exponentially in d,
inductive proofs that require d! terms, and
assumptions that imply a vanishing set of functions as d increases.
(Exercise: Think about how this applies to Monte Carlo methods.)
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 5 / 28
Philosophy
Naturally occuring functions of many variables are not general.
If a method can match what really occurs in some application,
then it can avoid the curse.
Non-trivial, non-circular characterizations of the set of functions that a
given method can match are hard. (I know of none.)
Instead we start from inspiration:
Neural networks are inspired by the visual cortex of cats.
The following method is inspired by partial differential equations in
physics (e.g. heat flow).
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 6 / 28
Approximation by Sums of Separable Tensors/Functions
In dimension d, a rank r approximation of a tensor T is
T(j1, j2, . . . , jd ) ≈ G(j1, . . . , jd ) =
r
l=1
d
i=1
Gl
i (ji )
or equivalently T ≈ G =
r
l=1
Gl
=
r
l=1
d
i=1
Gl
i .
Instead of $d , storage is rd$, which is no longer exponential.
To do functions, just change notation:
f (x1, x2, . . . , xd ) ≈ g(x1, . . . , xd ) =
r
l=1
d
i=1
gl
i (xi ) .
With large enough r this can approximate anything within .
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 7 / 28
Basic Computational Paradigm
1 Start with operators/matrices and functions/vectors that can be
represented within with low rank.
2 Do linear algebra operations with them, e.g.
˜g = Lg =
r
l=1
r1
m=1
d
i=1
(Ll
i gm
i (xi ))
The computational cost is O(d · r · r1)
which is linear in d rather than exponential.
3 Adaptively re-minimize the rank of the output of each operation,
controlling the approximation error.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 8 / 28
Example: Power Method
L =
r
g1 =
r1
multiply
↓
˜g =
r·r1
↓
reduce r · r1 → r2
L =
r
g2 =
r2
· · ·
↓
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 9 / 28
Reducing the Rank
We wish to (well) approximate ˜g =
R
m=1
d
i=1
˜gm
i by g =
r
l=1
d
i=1
gl
i ,
with r small(er).
This is NP-hard, but we can try optimization algorithms:
From an initial g, iteratively modify {gl
i } to reduce the error ˜g −g 2
2.
You can try your favorite generic method:
Newton’s method and variations
gradient descent and variations
GMRES, BFGS, other acronyms
etc.
Often any method will do, but sometimes all of them struggle.
(I have worked years on challenges with this optimization problem.)
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 10 / 28
Alternating Least Squares (ALS)
This optimization problem has a multilinear structure we can use.
Loop until the error is small enough or r seems insufficient:
Loop through the directions k = 1, . . . , d.
Fix {gl
i } for i = k, and solve a linear least squares problem for new gl
k .
The normal equations are




i=k g1
i , g1
i . . . i=k g1
i , gr
i
...
...
...
i=k gr
i , g1
i . . . i=k gr
i , gr
i







g1
k
...
gr
k


 =




R
q=1 ˜gR
k i=k g1
i , ˜gq
i
...
R
q=1 ˜gR
k i=k gr
i , ˜gq
i



 .
ALS is old, simple, stepwise robust, adaptable, and widely used,
but does not make the underlying optimization problem any easier.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 11 / 28
Extended Computational Paradigm
(developed mainly for quantum mechanics)
Some symmetries can be enforced implicitly in the inner product.
Example: The antisymmetrizer A creates the beast
A
N
i=1
φi (γi ) =
1
N!
φ1(γ1) φ1(γ2) · · · φ1(γN)
φ2(γ1) φ2(γ2) · · · φ2(γN)
...
...
...
φN(γ1) φN(γ2) · · · φN(γN)
,
but inner products with it are computed simply as
A ˜φi , A φi =
|L|
N!
with L(i, j) = ˜φi , φj .
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 12 / 28
Extended Computational Paradigm
If L does not have low rank but Lg1, g is computable,
then you cannot use the basic paradigm g1
apply L
−−−−→ Lg1 = ˜g
reduce rank
−−−−−−−→ g2
but you can sometimes still run ALS to form g.
Example: the electron-electron interaction (multiplication) operator
W =
1
2
N
i=1 j=i
1
ri − rj
cannot be written with small r, but
AW ˜φi , A φi
is computable (formula suppressed).
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 13 / 28
Extended Computational Paradigm
If you know why your function cannot be written will small r,
you might be able to extend the sum-of-separable format.
Example: To capture the interelectron cusp, we can use
A
P
p=0


1
2 m=n
wp(|γm − γn|)


rp
q=1
N
i=1
φp,q
i (γi ) .
Example: To scale to large systems (composed of subsystems) we can use
A
r
q=1
K
k=1


rk
qk =1
Nk
ik =1
φq,qk
k,ik
(γk,ik
)

 .
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 14 / 28
Conclusions, Part I
Sums of separable functions give a tractable way to represent (some)
functions of many variables.
You can compute with then, to solve PDEs etc.
There are various extensions.
(There are difficulties too, which I skip.)
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 15 / 28
Mutivariate Regression
Beginning with scattered data in high dimensions
D = (xj, yj) = (xj
1, · · · , xj
d ; yj)
N
j=1
,
define an empirical inner product between functions
f , g =
N
j=1
f (xj)g(xj) ,
which also works between a function and our data,
{(xj, yj)}N
j=1 , g =
N
j=1
yjg(xj) .
The (empirical) least-squares error is then
{(xj, yj)}N
j=1 − g
2
=
N
j=1
(yj − g(xj))2
.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 16 / 28
Regression with a Sum of Separable Functions
Construct g(x) such that g(xj) ≈ yj with
g(x) =
r
l=1
d
i=1
gl
i (xi ) .
We can use an ALS approach:
Loop until you are happy or the metaparameters seem inappropriate:
Loop through the directions k = 1, . . . , d.
Fix {gl
i } for i = k, and update {gl
k }l to reduce (minimize) the error
N
j=1
yj −
r
l=1
gl
k (xj
k )
d
i=k
gl
i (xj
i )
2
.
If we choose each gl
k to be a linear combination of some basis functions,
then we get a linear least-squares problem in its coefficients.
Otherwise (and for other loss functions) it is nonlinear.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 17 / 28
Comments
The usual issues (noise, local minima, over-fitting) and
standard techniques (regularization, cross-validation) apply.
The cost for an optimization pass is linear in both d and N,
so the method is feasible for large data sets in high dimensions.
As of 2009, this regression method was competitive on a standard set
of benchmark problems (see the paper).
As of 2010, a classification method based on these principles was
competitive on a standard set of benchmark problems (see a paper by
Jochen Garcke).
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 18 / 28
Regression on Molecules and Materials
D = {(σj, yj)}N
j=1, where σj is a material/molecular structure,
which is an unordered set of atoms a = (t, r),
where t is a species type (e.g. t = Mo), and
r is a location in 3-dimensional space.
A structure can be mapped to a set Vσ whose elements (w, v) are a
weight w and an ordered list of atoms v called a view.
The set Vσ is invariant under rotations, translations, and the order the
atoms are given in.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 19 / 28
E
T
  ©
rA rC
rB
rD
maps to
the views:
Weight a1 a2 a3 a4
1/4
E
T
  ©
rA E
T
  ©
rC E
T
  ©
r
B
E
T
  ©
rD
1/4
E
T
  ©
rB E
T
  ©
rC E
T
  ©
r
A
E
T
  ©
rD
1/8
E
T
  ©
rC E
T
  ©
rA E
T
  ©
rB
E
T
  ©
r
D
1/8
E
T
  ©
rC E
T
  ©
rB E
T
  ©
rA
E
T
  ©
r D
1/4
E
T
  ©
rD E
T
  ©
rC E
T
  ©
r
B
E
T
  ©
rA
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 20 / 28
Regression with Consistent Functions
From a function g on ordered lists of atoms, we can build a function on
structures that is rotation and translation invariant by defining
Cg(σ) =
(w,v)∈Vσ
wg(v) .
We can then attempt to minimize the least-squares error
D −Cg 2
=
1
N
N
j=1
(yj − Cg(σj))2
=
1
N
N
j=1


yj −
(w,v)∈Vσj
wg(v)



2
.
If g([a1, a2, . . .]) := g([a1, a2, . . . , ad ]) =
r
l=1
d
i=1
gl
i (ai ) ,
then ALS can be run. Each gl
i is a function of a = (t, r), so its domain is
several copies of R3, which is tractable.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 21 / 28
Conclusions, Part II
Sums of separable functions give a tractable way to represent (some)
functions of many variables.
You can do regression with then, for machine learning etc.
There are various extensions.
(There are difficulties too, which I skip.)
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 22 / 28
Examples: Gaussians and Radial Functions
a exp −b x 2
= a
d
i=1
exp −bx2
i
If φ(y) ≈
r
l=1
al e−bl y2
for 0 ≤ y , then
φ( x ) ≈
r
l=1
al exp −bl
d
i=1
x2
i =
r
l=1
al
d
i=1
exp −bl x2
i ,
with rank r independent of d (but be careful about ≈ when used).
This construction is especially useful for Greens functions such as 1/ r .
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 23 / 28
Example: Linear Model
If we can write
φ(t) ≈
r
l=1
αl exp(βl t)
then the linear model has
φ
d
i=1
ai xi + b ≈
r
l=1
αl exp βl
d
i=1
ai xi + b
=
r
l=1
αl exp(βl b)
d
i=1
exp(βl ai xi ) .
Properties of φ matter, but the orientation of the axes does not.
(Although if only one ai is nonzero, then r = 1.)
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 24 / 28
Example: Additive Model
f (x) =
d
i=1
fi (xi ) =
d
dt
d
i=1
(1 + tfi (xi ))
t=0
= lim
h→0
1
2h
d
i=1
(1 + hfi (xi )) −
d
i=1
(1 − hfi (xi )) .
At r = 2 the minimization problem is ill-posed.
Ill-posedness can allow useful approximations.
There can be large cancellations and ill-conditioning.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 25 / 28
Example: Sine of the sum of several variables
As long as sin(αk − αj) = 0 for all j = k,
sin


d
j=1
xj

 =
d
j=1
sin(xj)
d
k=1,k=j
sin(xk + αk − αj)
sin(αk − αj)
,
which is rank d.
Ordinary trigonometric expansions yield r = 2d .
Over the complex numbers, r = 2. The field matters.
The representation is not unique. (For generic tensors they are.)
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 26 / 28
Example: Do not add Constraints!
If {gj}2d
j=1 form an orthonormal set and
g(x) =
d
i=1
gi (xi ) +
d
i=1
(gi (xi ) + gi+d (xi ))
then an orthogonality constraint would force us to multiply out,
g(x) =
d
i=1
gi (xi ) +g1(x1)
d
i=2
(gi (xi ) + gi+d (xi ))
+g1+d (x1)
d
i=2
(gi (xi ) + gi+d (xi ))
= · · ·
and have r = 2d instead of r = 2.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 27 / 28
Final Thoughts
There are no theorems that this approach is good,
but there are intriguing examples.
There are not many alternatives for computing in high dimensions.
(There are alternative tensor formats.)
See http://www.ohiouniversityfaculty.com/mohlenka/
for papers.
Talk with me if any of this seems useful for you.
Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 28 / 28

More Related Content

What's hot

Reliable ABC model choice via random forests
Reliable ABC model choice via random forestsReliable ABC model choice via random forests
Reliable ABC model choice via random forestsChristian Robert
 
A method for finding an optimal solution of an assignment problem under mixed...
A method for finding an optimal solution of an assignment problem under mixed...A method for finding an optimal solution of an assignment problem under mixed...
A method for finding an optimal solution of an assignment problem under mixed...Navodaya Institute of Technology
 
from model uncertainty to ABC
from model uncertainty to ABCfrom model uncertainty to ABC
from model uncertainty to ABCChristian Robert
 
Eagle Strategy Using Levy Walk and Firefly Algorithms For Stochastic Optimiza...
Eagle Strategy Using Levy Walk and Firefly Algorithms For Stochastic Optimiza...Eagle Strategy Using Levy Walk and Firefly Algorithms For Stochastic Optimiza...
Eagle Strategy Using Levy Walk and Firefly Algorithms For Stochastic Optimiza...Xin-She Yang
 
Large-Scale Nonparametric Estimation of Vehicle Travel Time Distributions
Large-Scale Nonparametric Estimation of Vehicle Travel Time DistributionsLarge-Scale Nonparametric Estimation of Vehicle Travel Time Distributions
Large-Scale Nonparametric Estimation of Vehicle Travel Time DistributionsRikiya Takahashi
 
Uncertainty Awareness in Integrating Machine Learning and Game Theory
Uncertainty Awareness in Integrating Machine Learning and Game TheoryUncertainty Awareness in Integrating Machine Learning and Game Theory
Uncertainty Awareness in Integrating Machine Learning and Game TheoryRikiya Takahashi
 
A Tutorial of the EM-algorithm and Its Application to Outlier Detection
A Tutorial of the EM-algorithm and Its Application to Outlier DetectionA Tutorial of the EM-algorithm and Its Application to Outlier Detection
A Tutorial of the EM-algorithm and Its Application to Outlier DetectionKonkuk University, Korea
 
Likelihood free computational statistics
Likelihood free computational statisticsLikelihood free computational statistics
Likelihood free computational statisticsPierre Pudlo
 
Dr Chris Drovandi (QUT) - Bayesian Indirect Inference Using a Parametric Auxi...
Dr Chris Drovandi (QUT) - Bayesian Indirect Inference Using a Parametric Auxi...Dr Chris Drovandi (QUT) - Bayesian Indirect Inference Using a Parametric Auxi...
Dr Chris Drovandi (QUT) - Bayesian Indirect Inference Using a Parametric Auxi...QUT_SEF
 
Review of Metaheuristics and Generalized Evolutionary Walk Algorithm
Review of Metaheuristics and Generalized Evolutionary Walk AlgorithmReview of Metaheuristics and Generalized Evolutionary Walk Algorithm
Review of Metaheuristics and Generalized Evolutionary Walk AlgorithmXin-She Yang
 
GAN(と強化学習との関係)
GAN(と強化学習との関係)GAN(と強化学習との関係)
GAN(と強化学習との関係)Masahiro Suzuki
 
Equational axioms for probability calculus and modelling of Likelihood ratio ...
Equational axioms for probability calculus and modelling of Likelihood ratio ...Equational axioms for probability calculus and modelling of Likelihood ratio ...
Equational axioms for probability calculus and modelling of Likelihood ratio ...Advanced-Concepts-Team
 
RECENT ADVANCES in PREDICTIVE (MACHINE) LEARNING
RECENT ADVANCES in PREDICTIVE (MACHINE) LEARNINGRECENT ADVANCES in PREDICTIVE (MACHINE) LEARNING
RECENT ADVANCES in PREDICTIVE (MACHINE) LEARNINGbutest
 
Nature-Inspired Metaheuristic Algorithms
Nature-Inspired Metaheuristic AlgorithmsNature-Inspired Metaheuristic Algorithms
Nature-Inspired Metaheuristic AlgorithmsXin-She Yang
 
Be naive. not idiot.
Be naive. not idiot. Be naive. not idiot.
Be naive. not idiot. PredicSis
 
Principle of Maximum Entropy
Principle of Maximum EntropyPrinciple of Maximum Entropy
Principle of Maximum EntropyJiawang Liu
 
24 Machine Learning Combining Models - Ada Boost
24 Machine Learning Combining Models - Ada Boost24 Machine Learning Combining Models - Ada Boost
24 Machine Learning Combining Models - Ada BoostAndres Mendez-Vazquez
 

What's hot (20)

Reliable ABC model choice via random forests
Reliable ABC model choice via random forestsReliable ABC model choice via random forests
Reliable ABC model choice via random forests
 
A method for finding an optimal solution of an assignment problem under mixed...
A method for finding an optimal solution of an assignment problem under mixed...A method for finding an optimal solution of an assignment problem under mixed...
A method for finding an optimal solution of an assignment problem under mixed...
 
from model uncertainty to ABC
from model uncertainty to ABCfrom model uncertainty to ABC
from model uncertainty to ABC
 
Eagle Strategy Using Levy Walk and Firefly Algorithms For Stochastic Optimiza...
Eagle Strategy Using Levy Walk and Firefly Algorithms For Stochastic Optimiza...Eagle Strategy Using Levy Walk and Firefly Algorithms For Stochastic Optimiza...
Eagle Strategy Using Levy Walk and Firefly Algorithms For Stochastic Optimiza...
 
Large-Scale Nonparametric Estimation of Vehicle Travel Time Distributions
Large-Scale Nonparametric Estimation of Vehicle Travel Time DistributionsLarge-Scale Nonparametric Estimation of Vehicle Travel Time Distributions
Large-Scale Nonparametric Estimation of Vehicle Travel Time Distributions
 
02 math essentials
02 math essentials02 math essentials
02 math essentials
 
Uncertainty Awareness in Integrating Machine Learning and Game Theory
Uncertainty Awareness in Integrating Machine Learning and Game TheoryUncertainty Awareness in Integrating Machine Learning and Game Theory
Uncertainty Awareness in Integrating Machine Learning and Game Theory
 
A Tutorial of the EM-algorithm and Its Application to Outlier Detection
A Tutorial of the EM-algorithm and Its Application to Outlier DetectionA Tutorial of the EM-algorithm and Its Application to Outlier Detection
A Tutorial of the EM-algorithm and Its Application to Outlier Detection
 
Side 2019, part 1
Side 2019, part 1Side 2019, part 1
Side 2019, part 1
 
Likelihood free computational statistics
Likelihood free computational statisticsLikelihood free computational statistics
Likelihood free computational statistics
 
Dr Chris Drovandi (QUT) - Bayesian Indirect Inference Using a Parametric Auxi...
Dr Chris Drovandi (QUT) - Bayesian Indirect Inference Using a Parametric Auxi...Dr Chris Drovandi (QUT) - Bayesian Indirect Inference Using a Parametric Auxi...
Dr Chris Drovandi (QUT) - Bayesian Indirect Inference Using a Parametric Auxi...
 
Review of Metaheuristics and Generalized Evolutionary Walk Algorithm
Review of Metaheuristics and Generalized Evolutionary Walk AlgorithmReview of Metaheuristics and Generalized Evolutionary Walk Algorithm
Review of Metaheuristics and Generalized Evolutionary Walk Algorithm
 
GAN(と強化学習との関係)
GAN(と強化学習との関係)GAN(と強化学習との関係)
GAN(と強化学習との関係)
 
Equational axioms for probability calculus and modelling of Likelihood ratio ...
Equational axioms for probability calculus and modelling of Likelihood ratio ...Equational axioms for probability calculus and modelling of Likelihood ratio ...
Equational axioms for probability calculus and modelling of Likelihood ratio ...
 
RECENT ADVANCES in PREDICTIVE (MACHINE) LEARNING
RECENT ADVANCES in PREDICTIVE (MACHINE) LEARNINGRECENT ADVANCES in PREDICTIVE (MACHINE) LEARNING
RECENT ADVANCES in PREDICTIVE (MACHINE) LEARNING
 
Nature-Inspired Metaheuristic Algorithms
Nature-Inspired Metaheuristic AlgorithmsNature-Inspired Metaheuristic Algorithms
Nature-Inspired Metaheuristic Algorithms
 
Pareto Models, Slides EQUINEQ
Pareto Models, Slides EQUINEQPareto Models, Slides EQUINEQ
Pareto Models, Slides EQUINEQ
 
Be naive. not idiot.
Be naive. not idiot. Be naive. not idiot.
Be naive. not idiot.
 
Principle of Maximum Entropy
Principle of Maximum EntropyPrinciple of Maximum Entropy
Principle of Maximum Entropy
 
24 Machine Learning Combining Models - Ada Boost
24 Machine Learning Combining Models - Ada Boost24 Machine Learning Combining Models - Ada Boost
24 Machine Learning Combining Models - Ada Boost
 

Similar to 2019 Fall Series: Postdoc Seminars - Special Guest Lecture, Attacking the Curse of Dimensionality Using Sums of Separable Functions - Martin Mohlenkamp, September 11, 2019

A review of automatic differentiationand its efficient implementation
A review of automatic differentiationand its efficient implementationA review of automatic differentiationand its efficient implementation
A review of automatic differentiationand its efficient implementationssuserfa7e73
 
International Journal of Computer Science and Security Volume (4) Issue (2)
International Journal of Computer Science and Security Volume (4) Issue (2)International Journal of Computer Science and Security Volume (4) Issue (2)
International Journal of Computer Science and Security Volume (4) Issue (2)CSCJournals
 
Chap 8. Optimization for training deep models
Chap 8. Optimization for training deep modelsChap 8. Optimization for training deep models
Chap 8. Optimization for training deep modelsYoung-Geun Choi
 
Citython presentation
Citython presentationCitython presentation
Citython presentationAnkit Tewari
 
Towards Reinforcement Learning-based Aggregate Computing
Towards Reinforcement Learning-based Aggregate ComputingTowards Reinforcement Learning-based Aggregate Computing
Towards Reinforcement Learning-based Aggregate ComputingGianluca Aguzzi
 
Supervised Learning.pdf
Supervised Learning.pdfSupervised Learning.pdf
Supervised Learning.pdfgadissaassefa
 
GRADIENT OMISSIVE DESCENT IS A MINIMIZATION ALGORITHM
GRADIENT OMISSIVE DESCENT IS A MINIMIZATION ALGORITHMGRADIENT OMISSIVE DESCENT IS A MINIMIZATION ALGORITHM
GRADIENT OMISSIVE DESCENT IS A MINIMIZATION ALGORITHMijscai
 
Using Learning Automata in Coordination Among Heterogeneous Agents in a Compl...
Using Learning Automata in Coordination Among Heterogeneous Agents in a Compl...Using Learning Automata in Coordination Among Heterogeneous Agents in a Compl...
Using Learning Automata in Coordination Among Heterogeneous Agents in a Compl...Waqas Tariq
 
A NEW APPROACH IN DYNAMIC TRAVELING SALESMAN PROBLEM: A HYBRID OF ANT COLONY ...
A NEW APPROACH IN DYNAMIC TRAVELING SALESMAN PROBLEM: A HYBRID OF ANT COLONY ...A NEW APPROACH IN DYNAMIC TRAVELING SALESMAN PROBLEM: A HYBRID OF ANT COLONY ...
A NEW APPROACH IN DYNAMIC TRAVELING SALESMAN PROBLEM: A HYBRID OF ANT COLONY ...ijmpict
 
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...cscpconf
 
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...csandit
 
A simple framework for contrastive learning of visual representations
A simple framework for contrastive learning of visual representationsA simple framework for contrastive learning of visual representations
A simple framework for contrastive learning of visual representationsDevansh16
 
Asynchronous Stochastic Optimization, New Analysis and Algorithms
Asynchronous Stochastic Optimization, New Analysis and AlgorithmsAsynchronous Stochastic Optimization, New Analysis and Algorithms
Asynchronous Stochastic Optimization, New Analysis and AlgorithmsFabian Pedregosa
 
4 tracking objects of deformable shapes (1)
4 tracking objects of deformable shapes (1)4 tracking objects of deformable shapes (1)
4 tracking objects of deformable shapes (1)prj_publication
 
4 tracking objects of deformable shapes
4 tracking objects of deformable shapes4 tracking objects of deformable shapes
4 tracking objects of deformable shapesprj_publication
 

Similar to 2019 Fall Series: Postdoc Seminars - Special Guest Lecture, Attacking the Curse of Dimensionality Using Sums of Separable Functions - Martin Mohlenkamp, September 11, 2019 (20)

A review of automatic differentiationand its efficient implementation
A review of automatic differentiationand its efficient implementationA review of automatic differentiationand its efficient implementation
A review of automatic differentiationand its efficient implementation
 
International Journal of Computer Science and Security Volume (4) Issue (2)
International Journal of Computer Science and Security Volume (4) Issue (2)International Journal of Computer Science and Security Volume (4) Issue (2)
International Journal of Computer Science and Security Volume (4) Issue (2)
 
Chap 8. Optimization for training deep models
Chap 8. Optimization for training deep modelsChap 8. Optimization for training deep models
Chap 8. Optimization for training deep models
 
Citython presentation
Citython presentationCitython presentation
Citython presentation
 
Towards Reinforcement Learning-based Aggregate Computing
Towards Reinforcement Learning-based Aggregate ComputingTowards Reinforcement Learning-based Aggregate Computing
Towards Reinforcement Learning-based Aggregate Computing
 
Supervised Learning.pdf
Supervised Learning.pdfSupervised Learning.pdf
Supervised Learning.pdf
 
GRADIENT OMISSIVE DESCENT IS A MINIMIZATION ALGORITHM
GRADIENT OMISSIVE DESCENT IS A MINIMIZATION ALGORITHMGRADIENT OMISSIVE DESCENT IS A MINIMIZATION ALGORITHM
GRADIENT OMISSIVE DESCENT IS A MINIMIZATION ALGORITHM
 
Using Learning Automata in Coordination Among Heterogeneous Agents in a Compl...
Using Learning Automata in Coordination Among Heterogeneous Agents in a Compl...Using Learning Automata in Coordination Among Heterogeneous Agents in a Compl...
Using Learning Automata in Coordination Among Heterogeneous Agents in a Compl...
 
GDRR Opening Workshop - Modeling Approaches for High-Frequency Financial Time...
GDRR Opening Workshop - Modeling Approaches for High-Frequency Financial Time...GDRR Opening Workshop - Modeling Approaches for High-Frequency Financial Time...
GDRR Opening Workshop - Modeling Approaches for High-Frequency Financial Time...
 
An Evaluation of Models for Runtime Approximation in Link Discovery
An Evaluation of Models for Runtime Approximation in Link DiscoveryAn Evaluation of Models for Runtime Approximation in Link Discovery
An Evaluation of Models for Runtime Approximation in Link Discovery
 
A NEW APPROACH IN DYNAMIC TRAVELING SALESMAN PROBLEM: A HYBRID OF ANT COLONY ...
A NEW APPROACH IN DYNAMIC TRAVELING SALESMAN PROBLEM: A HYBRID OF ANT COLONY ...A NEW APPROACH IN DYNAMIC TRAVELING SALESMAN PROBLEM: A HYBRID OF ANT COLONY ...
A NEW APPROACH IN DYNAMIC TRAVELING SALESMAN PROBLEM: A HYBRID OF ANT COLONY ...
 
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
 
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
OPTIMIZATION IN ENGINE DESIGN VIA FORMAL CONCEPT ANALYSIS USING NEGATIVE ATTR...
 
A simple framework for contrastive learning of visual representations
A simple framework for contrastive learning of visual representationsA simple framework for contrastive learning of visual representations
A simple framework for contrastive learning of visual representations
 
Have you met Julia?
Have you met Julia?Have you met Julia?
Have you met Julia?
 
Gy3312241229
Gy3312241229Gy3312241229
Gy3312241229
 
IJCSI-2015-12-2-10138 (1) (2)
IJCSI-2015-12-2-10138 (1) (2)IJCSI-2015-12-2-10138 (1) (2)
IJCSI-2015-12-2-10138 (1) (2)
 
Asynchronous Stochastic Optimization, New Analysis and Algorithms
Asynchronous Stochastic Optimization, New Analysis and AlgorithmsAsynchronous Stochastic Optimization, New Analysis and Algorithms
Asynchronous Stochastic Optimization, New Analysis and Algorithms
 
4 tracking objects of deformable shapes (1)
4 tracking objects of deformable shapes (1)4 tracking objects of deformable shapes (1)
4 tracking objects of deformable shapes (1)
 
4 tracking objects of deformable shapes
4 tracking objects of deformable shapes4 tracking objects of deformable shapes
4 tracking objects of deformable shapes
 

More from The Statistical and Applied Mathematical Sciences Institute

More from The Statistical and Applied Mathematical Sciences Institute (20)

Causal Inference Opening Workshop - Latent Variable Models, Causal Inference,...
Causal Inference Opening Workshop - Latent Variable Models, Causal Inference,...Causal Inference Opening Workshop - Latent Variable Models, Causal Inference,...
Causal Inference Opening Workshop - Latent Variable Models, Causal Inference,...
 
2019 Fall Series: Special Guest Lecture - 0-1 Phase Transitions in High Dimen...
2019 Fall Series: Special Guest Lecture - 0-1 Phase Transitions in High Dimen...2019 Fall Series: Special Guest Lecture - 0-1 Phase Transitions in High Dimen...
2019 Fall Series: Special Guest Lecture - 0-1 Phase Transitions in High Dimen...
 
Causal Inference Opening Workshop - Causal Discovery in Neuroimaging Data - F...
Causal Inference Opening Workshop - Causal Discovery in Neuroimaging Data - F...Causal Inference Opening Workshop - Causal Discovery in Neuroimaging Data - F...
Causal Inference Opening Workshop - Causal Discovery in Neuroimaging Data - F...
 
Causal Inference Opening Workshop - Smooth Extensions to BART for Heterogeneo...
Causal Inference Opening Workshop - Smooth Extensions to BART for Heterogeneo...Causal Inference Opening Workshop - Smooth Extensions to BART for Heterogeneo...
Causal Inference Opening Workshop - Smooth Extensions to BART for Heterogeneo...
 
Causal Inference Opening Workshop - A Bracketing Relationship between Differe...
Causal Inference Opening Workshop - A Bracketing Relationship between Differe...Causal Inference Opening Workshop - A Bracketing Relationship between Differe...
Causal Inference Opening Workshop - A Bracketing Relationship between Differe...
 
Causal Inference Opening Workshop - Testing Weak Nulls in Matched Observation...
Causal Inference Opening Workshop - Testing Weak Nulls in Matched Observation...Causal Inference Opening Workshop - Testing Weak Nulls in Matched Observation...
Causal Inference Opening Workshop - Testing Weak Nulls in Matched Observation...
 
Causal Inference Opening Workshop - Difference-in-differences: more than meet...
Causal Inference Opening Workshop - Difference-in-differences: more than meet...Causal Inference Opening Workshop - Difference-in-differences: more than meet...
Causal Inference Opening Workshop - Difference-in-differences: more than meet...
 
Causal Inference Opening Workshop - New Statistical Learning Methods for Esti...
Causal Inference Opening Workshop - New Statistical Learning Methods for Esti...Causal Inference Opening Workshop - New Statistical Learning Methods for Esti...
Causal Inference Opening Workshop - New Statistical Learning Methods for Esti...
 
Causal Inference Opening Workshop - Bipartite Causal Inference with Interfere...
Causal Inference Opening Workshop - Bipartite Causal Inference with Interfere...Causal Inference Opening Workshop - Bipartite Causal Inference with Interfere...
Causal Inference Opening Workshop - Bipartite Causal Inference with Interfere...
 
Causal Inference Opening Workshop - Bridging the Gap Between Causal Literatur...
Causal Inference Opening Workshop - Bridging the Gap Between Causal Literatur...Causal Inference Opening Workshop - Bridging the Gap Between Causal Literatur...
Causal Inference Opening Workshop - Bridging the Gap Between Causal Literatur...
 
Causal Inference Opening Workshop - Some Applications of Reinforcement Learni...
Causal Inference Opening Workshop - Some Applications of Reinforcement Learni...Causal Inference Opening Workshop - Some Applications of Reinforcement Learni...
Causal Inference Opening Workshop - Some Applications of Reinforcement Learni...
 
Causal Inference Opening Workshop - Bracketing Bounds for Differences-in-Diff...
Causal Inference Opening Workshop - Bracketing Bounds for Differences-in-Diff...Causal Inference Opening Workshop - Bracketing Bounds for Differences-in-Diff...
Causal Inference Opening Workshop - Bracketing Bounds for Differences-in-Diff...
 
Causal Inference Opening Workshop - Assisting the Impact of State Polcies: Br...
Causal Inference Opening Workshop - Assisting the Impact of State Polcies: Br...Causal Inference Opening Workshop - Assisting the Impact of State Polcies: Br...
Causal Inference Opening Workshop - Assisting the Impact of State Polcies: Br...
 
Causal Inference Opening Workshop - Experimenting in Equilibrium - Stefan Wag...
Causal Inference Opening Workshop - Experimenting in Equilibrium - Stefan Wag...Causal Inference Opening Workshop - Experimenting in Equilibrium - Stefan Wag...
Causal Inference Opening Workshop - Experimenting in Equilibrium - Stefan Wag...
 
Causal Inference Opening Workshop - Targeted Learning for Causal Inference Ba...
Causal Inference Opening Workshop - Targeted Learning for Causal Inference Ba...Causal Inference Opening Workshop - Targeted Learning for Causal Inference Ba...
Causal Inference Opening Workshop - Targeted Learning for Causal Inference Ba...
 
Causal Inference Opening Workshop - Bayesian Nonparametric Models for Treatme...
Causal Inference Opening Workshop - Bayesian Nonparametric Models for Treatme...Causal Inference Opening Workshop - Bayesian Nonparametric Models for Treatme...
Causal Inference Opening Workshop - Bayesian Nonparametric Models for Treatme...
 
2019 Fall Series: Special Guest Lecture - Adversarial Risk Analysis of the Ge...
2019 Fall Series: Special Guest Lecture - Adversarial Risk Analysis of the Ge...2019 Fall Series: Special Guest Lecture - Adversarial Risk Analysis of the Ge...
2019 Fall Series: Special Guest Lecture - Adversarial Risk Analysis of the Ge...
 
2019 Fall Series: Professional Development, Writing Academic Papers…What Work...
2019 Fall Series: Professional Development, Writing Academic Papers…What Work...2019 Fall Series: Professional Development, Writing Academic Papers…What Work...
2019 Fall Series: Professional Development, Writing Academic Papers…What Work...
 
2019 GDRR: Blockchain Data Analytics - Machine Learning in/for Blockchain: Fu...
2019 GDRR: Blockchain Data Analytics - Machine Learning in/for Blockchain: Fu...2019 GDRR: Blockchain Data Analytics - Machine Learning in/for Blockchain: Fu...
2019 GDRR: Blockchain Data Analytics - Machine Learning in/for Blockchain: Fu...
 
2019 GDRR: Blockchain Data Analytics - QuTrack: Model Life Cycle Management f...
2019 GDRR: Blockchain Data Analytics - QuTrack: Model Life Cycle Management f...2019 GDRR: Blockchain Data Analytics - QuTrack: Model Life Cycle Management f...
2019 GDRR: Blockchain Data Analytics - QuTrack: Model Life Cycle Management f...
 

Recently uploaded

Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)eniolaolutunde
 
Interactive Powerpoint_How to Master effective communication
Interactive Powerpoint_How to Master effective communicationInteractive Powerpoint_How to Master effective communication
Interactive Powerpoint_How to Master effective communicationnomboosow
 
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPTECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPTiammrhaywood
 
The basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxThe basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxheathfieldcps1
 
call girls in Kamla Market (DELHI) 🔝 >༒9953330565🔝 genuine Escort Service 🔝✔️✔️
call girls in Kamla Market (DELHI) 🔝 >༒9953330565🔝 genuine Escort Service 🔝✔️✔️call girls in Kamla Market (DELHI) 🔝 >༒9953330565🔝 genuine Escort Service 🔝✔️✔️
call girls in Kamla Market (DELHI) 🔝 >༒9953330565🔝 genuine Escort Service 🔝✔️✔️9953056974 Low Rate Call Girls In Saket, Delhi NCR
 
URLs and Routing in the Odoo 17 Website App
URLs and Routing in the Odoo 17 Website AppURLs and Routing in the Odoo 17 Website App
URLs and Routing in the Odoo 17 Website AppCeline George
 
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...EduSkills OECD
 
Paris 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityParis 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityGeoBlogs
 
Hybridoma Technology ( Production , Purification , and Application )
Hybridoma Technology  ( Production , Purification , and Application  ) Hybridoma Technology  ( Production , Purification , and Application  )
Hybridoma Technology ( Production , Purification , and Application ) Sakshi Ghasle
 
Mastering the Unannounced Regulatory Inspection
Mastering the Unannounced Regulatory InspectionMastering the Unannounced Regulatory Inspection
Mastering the Unannounced Regulatory InspectionSafetyChain Software
 
Call Girls in Dwarka Mor Delhi Contact Us 9654467111
Call Girls in Dwarka Mor Delhi Contact Us 9654467111Call Girls in Dwarka Mor Delhi Contact Us 9654467111
Call Girls in Dwarka Mor Delhi Contact Us 9654467111Sapana Sha
 
Accessible design: Minimum effort, maximum impact
Accessible design: Minimum effort, maximum impactAccessible design: Minimum effort, maximum impact
Accessible design: Minimum effort, maximum impactdawncurless
 
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdfssuser54595a
 
Employee wellbeing at the workplace.pptx
Employee wellbeing at the workplace.pptxEmployee wellbeing at the workplace.pptx
Employee wellbeing at the workplace.pptxNirmalaLoungPoorunde1
 
MENTAL STATUS EXAMINATION format.docx
MENTAL     STATUS EXAMINATION format.docxMENTAL     STATUS EXAMINATION format.docx
MENTAL STATUS EXAMINATION format.docxPoojaSen20
 
Organic Name Reactions for the students and aspirants of Chemistry12th.pptx
Organic Name Reactions  for the students and aspirants of Chemistry12th.pptxOrganic Name Reactions  for the students and aspirants of Chemistry12th.pptx
Organic Name Reactions for the students and aspirants of Chemistry12th.pptxVS Mahajan Coaching Centre
 
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...Marc Dusseiller Dusjagr
 
Introduction to AI in Higher Education_draft.pptx
Introduction to AI in Higher Education_draft.pptxIntroduction to AI in Higher Education_draft.pptx
Introduction to AI in Higher Education_draft.pptxpboyjonauth
 
Alper Gobel In Media Res Media Component
Alper Gobel In Media Res Media ComponentAlper Gobel In Media Res Media Component
Alper Gobel In Media Res Media ComponentInMediaRes1
 

Recently uploaded (20)

Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)Software Engineering Methodologies (overview)
Software Engineering Methodologies (overview)
 
Interactive Powerpoint_How to Master effective communication
Interactive Powerpoint_How to Master effective communicationInteractive Powerpoint_How to Master effective communication
Interactive Powerpoint_How to Master effective communication
 
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPTECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
 
The basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptxThe basics of sentences session 2pptx copy.pptx
The basics of sentences session 2pptx copy.pptx
 
call girls in Kamla Market (DELHI) 🔝 >༒9953330565🔝 genuine Escort Service 🔝✔️✔️
call girls in Kamla Market (DELHI) 🔝 >༒9953330565🔝 genuine Escort Service 🔝✔️✔️call girls in Kamla Market (DELHI) 🔝 >༒9953330565🔝 genuine Escort Service 🔝✔️✔️
call girls in Kamla Market (DELHI) 🔝 >༒9953330565🔝 genuine Escort Service 🔝✔️✔️
 
URLs and Routing in the Odoo 17 Website App
URLs and Routing in the Odoo 17 Website AppURLs and Routing in the Odoo 17 Website App
URLs and Routing in the Odoo 17 Website App
 
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
Presentation by Andreas Schleicher Tackling the School Absenteeism Crisis 30 ...
 
Paris 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activityParis 2024 Olympic Geographies - an activity
Paris 2024 Olympic Geographies - an activity
 
Hybridoma Technology ( Production , Purification , and Application )
Hybridoma Technology  ( Production , Purification , and Application  ) Hybridoma Technology  ( Production , Purification , and Application  )
Hybridoma Technology ( Production , Purification , and Application )
 
Mastering the Unannounced Regulatory Inspection
Mastering the Unannounced Regulatory InspectionMastering the Unannounced Regulatory Inspection
Mastering the Unannounced Regulatory Inspection
 
Call Girls in Dwarka Mor Delhi Contact Us 9654467111
Call Girls in Dwarka Mor Delhi Contact Us 9654467111Call Girls in Dwarka Mor Delhi Contact Us 9654467111
Call Girls in Dwarka Mor Delhi Contact Us 9654467111
 
Accessible design: Minimum effort, maximum impact
Accessible design: Minimum effort, maximum impactAccessible design: Minimum effort, maximum impact
Accessible design: Minimum effort, maximum impact
 
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf
18-04-UA_REPORT_MEDIALITERAСY_INDEX-DM_23-1-final-eng.pdf
 
Employee wellbeing at the workplace.pptx
Employee wellbeing at the workplace.pptxEmployee wellbeing at the workplace.pptx
Employee wellbeing at the workplace.pptx
 
MENTAL STATUS EXAMINATION format.docx
MENTAL     STATUS EXAMINATION format.docxMENTAL     STATUS EXAMINATION format.docx
MENTAL STATUS EXAMINATION format.docx
 
Organic Name Reactions for the students and aspirants of Chemistry12th.pptx
Organic Name Reactions  for the students and aspirants of Chemistry12th.pptxOrganic Name Reactions  for the students and aspirants of Chemistry12th.pptx
Organic Name Reactions for the students and aspirants of Chemistry12th.pptx
 
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...
“Oh GOSH! Reflecting on Hackteria's Collaborative Practices in a Global Do-It...
 
Introduction to AI in Higher Education_draft.pptx
Introduction to AI in Higher Education_draft.pptxIntroduction to AI in Higher Education_draft.pptx
Introduction to AI in Higher Education_draft.pptx
 
Alper Gobel In Media Res Media Component
Alper Gobel In Media Res Media ComponentAlper Gobel In Media Res Media Component
Alper Gobel In Media Res Media Component
 
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
Model Call Girl in Tilak Nagar Delhi reach out to us at 🔝9953056974🔝
 

2019 Fall Series: Postdoc Seminars - Special Guest Lecture, Attacking the Curse of Dimensionality Using Sums of Separable Functions - Martin Mohlenkamp, September 11, 2019

  • 1. Attacking the Curse of Dimensionality using Sums of Separable Functions Martin J. Mohlenkamp Department of Mathematics http://www.ohiouniversityfaculty.com/mohlenka/ SAMSI, September 2019
  • 2. Abstract Naive computations involving a function of many variables suffer from the curse of dimensionality: the computational cost grows exponentially with the number of variables. One approach to bypassing the curse is to approximate the function as a sum of products of functions of one variable and compute in this format. When the variables are indices, a function of many variables is called a tensor, and this approach is to approximate and use the tensor in the (so-called) canonical tensor format. In this talk I will describe how such approximations can be used in numerical analysis and in machine learning. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 2 / 28
  • 3. Goals of this Talk Show you a tool that you may find useful. Hint at other things I know that you may find useful. Not Goals Convince you that this tool is better than other methods. Show that I am great. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 3 / 28
  • 4. The Curse of Dimensionality (discrete setting) d Name Notation Storage Visual 1 Vector vj $ 2 Matrix Ajk $$ 3 Tensor Tjkm $$$ > 3 Tensor T(j1, . . . , jd ) $d ? The cost to do anything, even store the object, grows exponentially in the dimension d. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 4 / 28
  • 5. The Curse of Dimensionality (function setting) To approximate a function f (x1, x2, . . . , xd ) that has smoothness p to accuracy costs −d/p = ( −1/p)d = $d . This curse is unavoidable for general function spaces (smoothness classes). If a method seems to avoid it, look for “constants” that grow exponentially in d, inductive proofs that require d! terms, and assumptions that imply a vanishing set of functions as d increases. (Exercise: Think about how this applies to Monte Carlo methods.) Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 5 / 28
  • 6. Philosophy Naturally occuring functions of many variables are not general. If a method can match what really occurs in some application, then it can avoid the curse. Non-trivial, non-circular characterizations of the set of functions that a given method can match are hard. (I know of none.) Instead we start from inspiration: Neural networks are inspired by the visual cortex of cats. The following method is inspired by partial differential equations in physics (e.g. heat flow). Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 6 / 28
  • 7. Approximation by Sums of Separable Tensors/Functions In dimension d, a rank r approximation of a tensor T is T(j1, j2, . . . , jd ) ≈ G(j1, . . . , jd ) = r l=1 d i=1 Gl i (ji ) or equivalently T ≈ G = r l=1 Gl = r l=1 d i=1 Gl i . Instead of $d , storage is rd$, which is no longer exponential. To do functions, just change notation: f (x1, x2, . . . , xd ) ≈ g(x1, . . . , xd ) = r l=1 d i=1 gl i (xi ) . With large enough r this can approximate anything within . Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 7 / 28
  • 8. Basic Computational Paradigm 1 Start with operators/matrices and functions/vectors that can be represented within with low rank. 2 Do linear algebra operations with them, e.g. ˜g = Lg = r l=1 r1 m=1 d i=1 (Ll i gm i (xi )) The computational cost is O(d · r · r1) which is linear in d rather than exponential. 3 Adaptively re-minimize the rank of the output of each operation, controlling the approximation error. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 8 / 28
  • 9. Example: Power Method L = r g1 = r1 multiply ↓ ˜g = r·r1 ↓ reduce r · r1 → r2 L = r g2 = r2 · · · ↓ Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 9 / 28
  • 10. Reducing the Rank We wish to (well) approximate ˜g = R m=1 d i=1 ˜gm i by g = r l=1 d i=1 gl i , with r small(er). This is NP-hard, but we can try optimization algorithms: From an initial g, iteratively modify {gl i } to reduce the error ˜g −g 2 2. You can try your favorite generic method: Newton’s method and variations gradient descent and variations GMRES, BFGS, other acronyms etc. Often any method will do, but sometimes all of them struggle. (I have worked years on challenges with this optimization problem.) Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 10 / 28
  • 11. Alternating Least Squares (ALS) This optimization problem has a multilinear structure we can use. Loop until the error is small enough or r seems insufficient: Loop through the directions k = 1, . . . , d. Fix {gl i } for i = k, and solve a linear least squares problem for new gl k . The normal equations are     i=k g1 i , g1 i . . . i=k g1 i , gr i ... ... ... i=k gr i , g1 i . . . i=k gr i , gr i        g1 k ... gr k    =     R q=1 ˜gR k i=k g1 i , ˜gq i ... R q=1 ˜gR k i=k gr i , ˜gq i     . ALS is old, simple, stepwise robust, adaptable, and widely used, but does not make the underlying optimization problem any easier. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 11 / 28
  • 12. Extended Computational Paradigm (developed mainly for quantum mechanics) Some symmetries can be enforced implicitly in the inner product. Example: The antisymmetrizer A creates the beast A N i=1 φi (γi ) = 1 N! φ1(γ1) φ1(γ2) · · · φ1(γN) φ2(γ1) φ2(γ2) · · · φ2(γN) ... ... ... φN(γ1) φN(γ2) · · · φN(γN) , but inner products with it are computed simply as A ˜φi , A φi = |L| N! with L(i, j) = ˜φi , φj . Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 12 / 28
  • 13. Extended Computational Paradigm If L does not have low rank but Lg1, g is computable, then you cannot use the basic paradigm g1 apply L −−−−→ Lg1 = ˜g reduce rank −−−−−−−→ g2 but you can sometimes still run ALS to form g. Example: the electron-electron interaction (multiplication) operator W = 1 2 N i=1 j=i 1 ri − rj cannot be written with small r, but AW ˜φi , A φi is computable (formula suppressed). Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 13 / 28
  • 14. Extended Computational Paradigm If you know why your function cannot be written will small r, you might be able to extend the sum-of-separable format. Example: To capture the interelectron cusp, we can use A P p=0   1 2 m=n wp(|γm − γn|)   rp q=1 N i=1 φp,q i (γi ) . Example: To scale to large systems (composed of subsystems) we can use A r q=1 K k=1   rk qk =1 Nk ik =1 φq,qk k,ik (γk,ik )   . Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 14 / 28
  • 15. Conclusions, Part I Sums of separable functions give a tractable way to represent (some) functions of many variables. You can compute with then, to solve PDEs etc. There are various extensions. (There are difficulties too, which I skip.) Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 15 / 28
  • 16. Mutivariate Regression Beginning with scattered data in high dimensions D = (xj, yj) = (xj 1, · · · , xj d ; yj) N j=1 , define an empirical inner product between functions f , g = N j=1 f (xj)g(xj) , which also works between a function and our data, {(xj, yj)}N j=1 , g = N j=1 yjg(xj) . The (empirical) least-squares error is then {(xj, yj)}N j=1 − g 2 = N j=1 (yj − g(xj))2 . Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 16 / 28
  • 17. Regression with a Sum of Separable Functions Construct g(x) such that g(xj) ≈ yj with g(x) = r l=1 d i=1 gl i (xi ) . We can use an ALS approach: Loop until you are happy or the metaparameters seem inappropriate: Loop through the directions k = 1, . . . , d. Fix {gl i } for i = k, and update {gl k }l to reduce (minimize) the error N j=1 yj − r l=1 gl k (xj k ) d i=k gl i (xj i ) 2 . If we choose each gl k to be a linear combination of some basis functions, then we get a linear least-squares problem in its coefficients. Otherwise (and for other loss functions) it is nonlinear. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 17 / 28
  • 18. Comments The usual issues (noise, local minima, over-fitting) and standard techniques (regularization, cross-validation) apply. The cost for an optimization pass is linear in both d and N, so the method is feasible for large data sets in high dimensions. As of 2009, this regression method was competitive on a standard set of benchmark problems (see the paper). As of 2010, a classification method based on these principles was competitive on a standard set of benchmark problems (see a paper by Jochen Garcke). Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 18 / 28
  • 19. Regression on Molecules and Materials D = {(σj, yj)}N j=1, where σj is a material/molecular structure, which is an unordered set of atoms a = (t, r), where t is a species type (e.g. t = Mo), and r is a location in 3-dimensional space. A structure can be mapped to a set Vσ whose elements (w, v) are a weight w and an ordered list of atoms v called a view. The set Vσ is invariant under rotations, translations, and the order the atoms are given in. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 19 / 28
  • 20. E T   © rA rC rB rD maps to the views: Weight a1 a2 a3 a4 1/4 E T   © rA E T   © rC E T   © r B E T   © rD 1/4 E T   © rB E T   © rC E T   © r A E T   © rD 1/8 E T   © rC E T   © rA E T   © rB E T   © r D 1/8 E T   © rC E T   © rB E T   © rA E T   © r D 1/4 E T   © rD E T   © rC E T   © r B E T   © rA Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 20 / 28
  • 21. Regression with Consistent Functions From a function g on ordered lists of atoms, we can build a function on structures that is rotation and translation invariant by defining Cg(σ) = (w,v)∈Vσ wg(v) . We can then attempt to minimize the least-squares error D −Cg 2 = 1 N N j=1 (yj − Cg(σj))2 = 1 N N j=1   yj − (w,v)∈Vσj wg(v)    2 . If g([a1, a2, . . .]) := g([a1, a2, . . . , ad ]) = r l=1 d i=1 gl i (ai ) , then ALS can be run. Each gl i is a function of a = (t, r), so its domain is several copies of R3, which is tractable. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 21 / 28
  • 22. Conclusions, Part II Sums of separable functions give a tractable way to represent (some) functions of many variables. You can do regression with then, for machine learning etc. There are various extensions. (There are difficulties too, which I skip.) Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 22 / 28
  • 23. Examples: Gaussians and Radial Functions a exp −b x 2 = a d i=1 exp −bx2 i If φ(y) ≈ r l=1 al e−bl y2 for 0 ≤ y , then φ( x ) ≈ r l=1 al exp −bl d i=1 x2 i = r l=1 al d i=1 exp −bl x2 i , with rank r independent of d (but be careful about ≈ when used). This construction is especially useful for Greens functions such as 1/ r . Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 23 / 28
  • 24. Example: Linear Model If we can write φ(t) ≈ r l=1 αl exp(βl t) then the linear model has φ d i=1 ai xi + b ≈ r l=1 αl exp βl d i=1 ai xi + b = r l=1 αl exp(βl b) d i=1 exp(βl ai xi ) . Properties of φ matter, but the orientation of the axes does not. (Although if only one ai is nonzero, then r = 1.) Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 24 / 28
  • 25. Example: Additive Model f (x) = d i=1 fi (xi ) = d dt d i=1 (1 + tfi (xi )) t=0 = lim h→0 1 2h d i=1 (1 + hfi (xi )) − d i=1 (1 − hfi (xi )) . At r = 2 the minimization problem is ill-posed. Ill-posedness can allow useful approximations. There can be large cancellations and ill-conditioning. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 25 / 28
  • 26. Example: Sine of the sum of several variables As long as sin(αk − αj) = 0 for all j = k, sin   d j=1 xj   = d j=1 sin(xj) d k=1,k=j sin(xk + αk − αj) sin(αk − αj) , which is rank d. Ordinary trigonometric expansions yield r = 2d . Over the complex numbers, r = 2. The field matters. The representation is not unique. (For generic tensors they are.) Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 26 / 28
  • 27. Example: Do not add Constraints! If {gj}2d j=1 form an orthonormal set and g(x) = d i=1 gi (xi ) + d i=1 (gi (xi ) + gi+d (xi )) then an orthogonality constraint would force us to multiply out, g(x) = d i=1 gi (xi ) +g1(x1) d i=2 (gi (xi ) + gi+d (xi )) +g1+d (x1) d i=2 (gi (xi ) + gi+d (xi )) = · · · and have r = 2d instead of r = 2. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 27 / 28
  • 28. Final Thoughts There are no theorems that this approach is good, but there are intriguing examples. There are not many alternatives for computing in high dimensions. (There are alternative tensor formats.) See http://www.ohiouniversityfaculty.com/mohlenka/ for papers. Talk with me if any of this seems useful for you. Martin J. Mohlenkamp (OHIO) Attacking the CoD using SoS Functions SAMSI, September 2019 28 / 28