Upload
others
View
8
Download
0
Embed Size (px)
Citation preview
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Prof. Dr. Michael FeindtCETAInstitut für Experimentelle KernphysikUniversität Karlsruhe
Universität Freiburg, 23. November 2005
Physics with Neural Networks
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
ContentsHypothesis testing / classificationNeural networks
Classification, conditional probability densitiesNeuroBayes (robust regularised Bayesian neural network)Applications in DELPHI, CDF II, outside physics
Unusual but powerful network applications: Training with MC/data mixturesTraining on data onlyTraining with weightsTraining with background subtraction Multidimensional regression using NeuroBayes
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Hypothesis testing
Cut in a test-statistic:Accept hypothesis H0, if t<t(cut)
Error of 1. kind:P1(true hypothesis will be rejected)
Error of 2. kind:P2(wrong hypothesis is accepted)
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Hypothesis testing
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Hypothesis testing
A statistical method is thebetter the nearer it reachesthe point (1,1) in thepurity-efficiency-plot
Optimal choice of workingpoint according to particulartask: How does the total errorof the analysis scale withε und P?
0.7 0.8 0.9 1.0
Optimalworking point
Different cuts in t
Different cuts in t
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Determining the working point (scan through cuts on network output)
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Construction of a test statistic:How to make 100 dimensions one real number…
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Neural Networks
Neural networks:Self learning procedures, copied from nature
ParietalCortexFrontal Lobe
Motor Cortex
Temporal Lobe
Brain Stem
OccipitalLobe
Cerebellum
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Neural Networks
The information(the knowledge, the expertise)is coded in the connectionsbetween the neurons
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Neural networks
Basis function
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Neural networks – transfer functions
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Neural networks - training
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
f t
Probability that hypothesisis correct (classification)or probability densityfor variable t
t
How it works: training and applicationHistoric or simulateddata
Data seta = ...b = ...c = .......t = …!
NeuroBayesNeuroBayes®®TeacherTeacher
NeuroBayesNeuroBayes®®ExpertExpert
Actual (new real) data
Data seta = ...b = ...c = .......t = ?
ExpertiseExpertise
Expert system
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Naïve networks and criticizmWe‘ve tried that but it didn‘t give good results
- Stuck in local minimum- Learning not robust
We‘ve tried that but it was worse than our 100 person-yearsanalytical high tech algorithm
- Selected too naive input variables- Use your fance algorithm as INPUT !
We‘ve tried that but the predictions were wrong- Overtraining: the net learnt statistical fluctuations
Yeah but how can you estimate systematic errors?- How can you with cuts when variables are correlated?
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Address all these topics and build a professional robust and flexible NeuralNetwork package for physics, insurance, bank and industry applications:NeuroBayes®
<phi-t>: Foundation out of University of Karlsruhe, sponsored by exist-seed-programme of thefederal ministery for Education and ResearchBMBF
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
2000-2002 NeuroBayes®-specialisationfor economy at the University of Karlsruhe
Oct. 2002: GmbH founded, first industrial projects
June 2003: Removal into new office199 qm IT-Portal Karlsruhe
Exclusive rights for NeuroBayes®
Personell currently4 full time staff (all from HEP) anda number of associated people,Prof. consultance z.B. by Prof. Dr. Volker Blobel,Economic/legal/marketing- expertise present
Customers: Badische Versicherungen, dm drogerie markt, Otto Versand
History
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
NeuroBayes® principle
NeuroBayes® Teacher:Learning of complex relationships from existingdata bases
NeuroBayes® Expert:Prognosis for unknown data
Output
Input
Sign
ific
ance
cont
rol
Postprocessing
Preprocessing
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Bayesian RegularisationUse Bayesian arguments to regularize network learning:
PosteriorPosterior EvidenceEvidence
LikelihoodLikelihood PriorPrior
Learn only statistically relevant information, suppress statistical noise
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Ramler-plot (extended correlation matrix)
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Ramler-II-plot (visualize correlation to target)
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Visualisation of single input-variables
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Visualisation of correlation matrix
Variable 1: Training target
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Visualisation of network performance
Purity vs. efficiency
Signal-effiziency vs. total efficiency(Lift chart)
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Visualisation of NeuroBayes network Topology
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Example DELPHI kaon ID ClassificationClassificationIdentification of kaons:
For Phi reconstruction:For Phi reconstruction:Efficiency twice as good Efficiency twice as good at the same background at the same background level (i.e. S/B doubled)level (i.e. S/B doubled)
Green: better of two Green: better of two competing standard competing standard algorithms (>50 men algorithms (>50 men years)years)
NeuroBayesNeuroBayes®®::Finds who is right under Finds who is right under which circumstanceswhich circumstances
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
NeuroBayes:Increase of efficiencyat same purity as best cut basedselection:Improve significancefrom 18.4 to 22.1
NeuroBayes:Increase of efficiencyat same purity as best cut basedselection:Improve significancefrom 18.4 to 22.1
Optimisation of Bs reconstruction for Bs oscillation analysisOptimisation of Bs reconstruction for Bs oscillation analysis
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Neural Nets for B flavour taggingNeural Nets for B flavour tagging
NeuroBayes track net:Does a track come from a B-decay or not?Tracknet 1 – Vertexnet – Tracknet 2
NeuroBayes track net:Does a track come from a B-decay or not?Tracknet 1 – Vertexnet – Tracknet 2
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Aim: Bayesian estimator for a singlemultidimensional measurement .
"Components of may be correlated."Components of should be correlated to t or its uncertainty. "All this should be learned automatically in a robust way from data bases"containing Monte-Carlo simulations or historical data.
Aim: Bayesian estimator for a singlemultidimensional measurement .
"Components of may be correlated."Components of should be correlated to t or its uncertainty. "All this should be learned automatically in a robust way from data bases"containing Monte-Carlo simulations or historical data.
Note: Conditional probability density contains much more information than just the mean value, which is determined in a regression analysis. It also tells us something about the uncertainty and the form of thedistribution, in particular non-Gaussian tails.
Note: Conditional probability density contains much more information than just the mean value, which is determined in a regression analysis. It also tells us something about the uncertainty and the form of thedistribution, in particular non-Gaussian tails.
xr
xr
xr
)|( xtfr
ConditionalConditional probabilityprobability densitydensity reconstructionreconstruction::
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Conditional probability densitiesin particle physics
What is the probability density of the true B energy in this event taken with the DELPHI detector at LEP IIat this beam energy, this effective c.m. energythese n tracks with those momenta and rapidities in the hemisphere, which are forming this secondary vertex with this decay length and probability, this number of not well reconstructed tracks, this neutral showers,etc pp )|( xtf
r
t
xr
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Conditional probability densitiesin investment bankingWhat is the probability density for a price change of equity A in the neyt 10 days…
that made this and that price movement in thelast days and weeks…
is so much more expensive than the n-days moving average…
but is so much less expensive that the absolute maximum…
has this correlation to the crude oil price…and the Dow Jones index…and etc. pp.
)|( xtfr
t
xr
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Prediction of the completeprobability distributionwith the < phi-t > NeuroBayes neural network
t
)|( xtfr
ModeExpectation value
Standard deviationvolatility
Deviations fromnormal distribution,e.g. crash probability
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Conditional probability densities f(t|x)
Conditional probability densities f(t|x) are functions of x, but also depend on marginal distribution f(t).
Conditional probability densities f(t|x) are functions of x, but also depend on marginal distribution f(t).
Inclusive distribution(Bayesian Prior)
Inclusive distribution(Bayesian Prior)
Conditional probability density for a special case x
(Bayesian Posterior)
Conditional probability density for a special case x
(Bayesian Posterior)
Marginal distribution f(t)Marginal distribution f(t)
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Classical ansatz:f(x|t)=f(t|x)
approximately correctat good resolution
far away fromphysical boundaries
Bayesian ansatz:takes into accounta priori- knowledge f(t):•Lifetime never negative•True lifetime exponentiallydistributed
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
No selection:Improved resolution
No selection:Improved resolution
NeuroBayes phi-direction
Best ”classical" chi**2- fit
(BSAURUS)
NeuroBayes phi-direction
Best ”classical" chi**2- fit
(BSAURUS)
Resolution of azimuthal angle of inclusively reconstructed B-hadrons in the DELPHI-
detector
first neural reconstruction of a direction
Resolution of azimuthal angle of inclusively reconstructed B-hadrons in the DELPHI-
detector
first neural reconstruction of a direction
Direction of B-mesons (DELPHI)
After selection cut on estimated error:Resolution massively improved, no tails
==> allows reliable selection of good events
After selection cut on estimated error:Resolution massively improved, no tails
==> allows reliable selection of good events
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Some applications in high energy physicsDELPHI:Kaon, proton, electron idOptimisation of resolutions inclusive B- E, phi, Theta, Q-valueB** enrichmentB fragmentation functionLimit on B_s-mixingB0-mixingB- F/B-asymmetryB-> wrong sign charmCDF: (Work in Progress)Electron ID, muon ID, kaon/proton IDOptimisation of resonance reconstruction (X, Y , B_s) Spin parity analysisB-Tagging for top, Higgs, etc. B-Flavour Tagging for mixing analyses
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Applications of NeuroBayes in Economy> Medicine and Pharma research
e.g. effects and undesirable effects of drugsearly tumor recognition
> Banks e.g. Credit-Scoring (Basel II), Finance time seriesprediction, valuation of derivates, risk minimisedtrading strategies, client valuation
> Insurancese.g. risk and cost prediction for individual clients, probability of contract cancellation, fraud recognition, justice in tariffs
> Trading chain stores: turnover prognosis for individual articles/stores
Necessary prerequisite:Historic or simulated data must be available!
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Risk analysis for a car insuranceBGV
Results for a the Badischen Gemeinde-Versicherungen:
since May 2003: radically new tariff for young drivers!
New variables added to calculation of the premium.Correlations taken into account.
Risk und premium up to a factor of 3 apart from each other!Even probability distribution of height of can be predicted
Premature contract cancellation also well predictable
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
The ‘‘unjustice‘‘ of insurance premiums
The majority of customers (with lowrisk) are paying too much.
Less than half of the customers (withlarger risk) do not pay enough, someby far not enough.These are currently subsidised by themore careful customers.
Anza
hl K
unde
n
Ratio of the accident risk calculated using NeuroBayes®to premium paid (normalised to same total premium sum):
Risiko/Prämie
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Prediction of contract cancellationfor an insurance
The predictionreally holds:
Test on a a newstatistic year
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Bank customers contract cancellation
Prognosis of prematurecontract cancellation
Massive enrichment possible.
CRM measures introduced
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Near Future TurnaroundPredictions for Chain Stores
1. Time series modelling2. Correction and error estimate
using NeuroBayes
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
The <phi-t> mouse game:or:
even your ``free will´´ is predictable
www.phi-t.de/mousegame
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Direct competition…Data-Mining-Cup 2005:Task: Internet shop: Prognosis of customerswho will not pay
531 Participantsfrom 176 Universitiesfrom 41 countries.
6 Karlsruhe studentsall using Phi-TNeuroBayes®:TOP RANKINGS: Positions 2,3,4,5,6,7!
www.data-mining-cup.de
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
How does NeuroBayes compare to other neural networks?How does NeuroBayes compare to other neural networks?
depends on problem.Usually superiorbecause of robustpreprocessing and Bayesian regularisation.
(does not meanthat Jetnet couldnot also do it ifpreprocessing isdone by hand)
depends on problem.Usually superiorbecause of robustpreprocessing and Bayesian regularisation.
(does not meanthat Jetnet couldnot also do it ifpreprocessing isdone by hand)Example: electron ID for CDF IIExample: electron ID for CDF II
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Hadron collider: No good MC for backgrounds availableMC for resonance production with different JPC assumptions
Idea: take background from sidebands in datacheck that network cannot learn mass
Hadron collider: No good MC for backgrounds availableMC for resonance production with different JPC assumptions
Idea: take background from sidebands in datacheck that network cannot learn mass
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
background-likehardly loose any signalbackground-likehardly loose any signal
signal-likesignal-like
X
soft NeuroBayesselection
X(3872) analysis:X(3872) analysis:
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Hard NeuroBayes cut:Very clean X(3872) signalHard NeuroBayes cut:Very clean X(3872) signal
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Training with weights:Neural Network Spin Parity AnalysisTraining with weights:Neural Network Spin Parity Analysis
Use data from sidebands as background sample
Use phase space decay MC with modelled pT distributionas signal sample
Calculate squared amplitudes for specific spin-parity assignmentsand use as weights in training
Hard cut on neural network trained by correct hypothesis shouldincrease signal over background
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Very hard NeuroBayes cuts:Very hard NeuroBayes cuts:
A good hypothesis for X A good hypothesis for X
JPC=1—(ππ)s hypothesisJPC=1—(ππ)s hypothesis
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Very hard NeuroBayes cuts:Very hard NeuroBayes cuts:
a not so good hypothesis for Xa not so good hypothesis for X
JPC=1—(ππ)s hypothesisJPC=1—(ππ)s hypothesis
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Hadron collider: Fast resonance S/N optimisation without MC:
Idea: Training with background subtractionSignal: Peak region weight 1
Sideband region with weight -1Background: Sideband region with weight 1
Hadron collider: Fast resonance S/N optimisation without MC:
Idea: Training with background subtractionSignal: Peak region weight 1
Sideband region with weight -1Background: Sideband region with weight 1
works very well!
also for Y(2S)and Y(3S) !Although just trained on Y(1S)
works very well!
also for Y(2S)and Y(3S) !Although just trained on Y(1S)
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
works very well also for Y(2S) and Y(3S) !Really learn to separate resonance from combinatorial backgroundworks very well also for Y(2S) and Y(3S) !Really learn to separate resonance from combinatorial background
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Making MC for hadronic background without specific model:Multidimensional correlated regression using NeuroBayesMaking MC for hadronic background without specific model:Multidimensional correlated regression using NeuroBayes
Use data in non-resonance region as signalUse phase space MC as background
Train NeuroBayes network,NN output O is Bayesian a posteriori probability that event stemsfrom signal (i.e. data distribution) rather than phase space MC:
O=P(S) with P(S)+P(B)=1Calculate weight W= P(S)/P(B) = O/(1-O) Phase space MC events with this weight W look like data!MC modelling of complicated background is possible!Opens new roads for likelihood fits
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005
Some kinematical variableDistributions(J/psi pi+pi- selection)
Black: real dataRed: weighted phase space MC
Physics with Neural Networks Michael Feindt Seminar Uni Freiburg Nov. 2005