A new approach for the gamma tracking clustering by the deterministic annealing method François Didierjean IPHC, Strasbourg.

Slides:



Advertisements
Similar presentations
Feature Selection as Relevant Information Encoding Naftali Tishby School of Computer Science and Engineering The Hebrew University, Jerusalem, Israel NIPS.
Advertisements

Advanced GAmma Tracking Array
Clustering Clustering of data is a method by which large sets of data is grouped into clusters of smaller sets of similar data. The example below demonstrates.
CHAPTER 8 A NNEALING- T YPE A LGORITHMS Organization of chapter in ISSO –Introduction to simulated annealing –Simulated annealing algorithm Basic algorithm.
Chapter 3 The Greedy Method 3.
AGATA Introduction John Simpson Nuclear Physics Group.
Michel Lefebvre, 2007/03/02Optimal Jet Finder1 Optimal Jet Finder work in progress Physics and Astronomy University of Victoria British Columbia, Canada.
The canonical ensemble System  Heat Reservoir R T=const. adiabatic wall Consider system at constant temperature and volume We have shown in thermodynamics.
Geant4 simulations for the calorimeter prototypes D. Di Julio, J. Cederkäll, P. Golubev, B. Jakobsson Lund University, Lund, Sweden.
2007/03/20Optimal Jet Finder1 Physics and Astronomy University of Victoria British Columbia, Canada Damir Lelas Rolf Seuster Michel Lefebvre, Rob McPherson,
Clustering… in General In vector space, clusters are vectors found within  of a cluster vector, with different techniques for determining the cluster.
MAE 552 – Heuristic Optimization Lecture 8 February 8, 2002.
Geant4 simulations for the calorimeter prototypes D. Di Julio, J. Cederkäll, P. Golubev, B. Jakobsson Lund University, Lund, Sweden.
MAE 552 – Heuristic Optimization Lecture 6 February 6, 2002.
1 Approximate Solution to an Exam Timetabling Problem Adam White Dept of Computing Science University of Alberta Adam White Dept of Computing Science University.
Prénom Nom Document Analysis: Data Analysis and Clustering Prof. Rolf Ingold, University of Fribourg Master course, spring semester 2008.
1 Power Control, Interference Suppression and Interference Avoidance in Wireless Systems Roy Yates (with S. Ulukus and C. Rose) WINLAB, Rutgers University.
Zoë Abrams, Ashish Goel, Serge Plotkin Stanford University Set K-Cover Algorithms for Energy Efficient Monitoring in Wireless Sensor Networks.
Cluster Analysis (1).
Preference Analysis Joachim Giesen and Eva Schuberth May 24, 2006.
Simulations with MEGAlib Jau-Shian Liang Department of Physics, NTHU / SSL, UCB 2007/05/15.
Clustering with Bregman Divergences Arindam Banerjee, Srujana Merugu, Inderjit S. Dhillon, Joydeep Ghosh Presented by Rohit Gupta CSci 8980: Machine Learning.
Clustering Ram Akella Lecture 6 February 23, & 280I University of California Berkeley Silicon Valley Center/SC.
AGATA Demonstrator Test With a 252 Cf Source: Neutron-Gamma Discrimination Menekşe ŞENYİĞİT.
Workshop on Physics on Nuclei at Extremes, Tokyo Institute of Technology, Institute for Nuclear Research and Nuclear Energy Bulgarian Academy.
1 IE 607 Heuristic Optimization Simulated Annealing.
Modeling and simulation of systems Simulation optimization and example of its usage in flexible production system control.
1 On the importance of nucleation for the formation of quark cores inside compact stars Bruno Werneck Mintz* Eduardo Souza Fraga Universidade Federal do.
Heat Capacities of 56 Fe and 57 Fe Emel Algin Eskisehir Osmangazi University Workshop on Level Density and Gamma Strength in Continuum May 21-24, 2007.
Generative Topographic Mapping in Life Science Jong Youl Choi School of Informatics and Computing Pervasive Technology Institute Indiana University
Generative Topographic Mapping by Deterministic Annealing Jong Youl Choi, Judy Qiu, Marlon Pierce, and Geoffrey Fox School of Informatics and Computing.
Boltzmann Machine (BM) (§6.4) Hopfield model + hidden nodes + simulated annealing BM Architecture –a set of visible nodes: nodes can be accessed from outside.
Simulated Annealing.
Summary of PHOS Internal Notes (part I) Rafael Diaz Valdes 10/25/20151.
Doshisha Univ., Kyoto, Japan CEC2003 Adaptive Temperature Schedule Determined by Genetic Algorithm for Parallel Simulated Annealing Doshisha University,
Markov Chain Monte Carlo and Gibbs Sampling Vasileios Hatzivassiloglou University of Texas at Dallas.
A New Analytic Model for the Production of X-ray Time Lags in Radio Loud AGN and X-Ray Binaries John J. Kroon Peter A. Becker George Mason University MARLAM.
Concept of the scanning table in Strasbourg François DIDIERJEAN Tatjana FAUL, Fabrice STEHLIN Strasbourg AGATA week July 2008 Uppsala, Sweden.
The Information Bottleneck Method clusters the response space, Y, into a much smaller space, T. In order to informatively cluster the response space, the.
Simulation of the energy response of  rays in CsI crystal arrays Thomas ZERGUERRAS EXL-R3B Collaboration Meeting, Orsay (France), 02/02/ /03/2006.
What is in my contribution area Nick Sinev, University of Oregon.
Simulated Annealing G.Anuradha.
July 11, 2006Bayesian Inference and Maximum Entropy Probing the covariance matrix Kenneth M. Hanson T-16, Nuclear Physics; Theoretical Division Los.
A split-and-merge framework for 2D shape summarization D. Gerogiannis, C. Nikou and A. Likas Department of Computer Science, University of Ioannina, Greece.
Genotype Calling Matt Schuerman. Biological Problem How do we know an individual’s SNP values (genotype)? Each SNP can have two values (A/B) Each individual.
Tracking Background GRETINA Software Working Group Meeting September 21-22, 2012, NSCL MSU I-Yang Lee Lawrence Berkeley National Laboratory.
Ramakrishna Lecture#2 CAD for VLSI Ramakrishna
T. Lari – INFN Milan Status of ATLAS Pixel Test beam simulation Status of the validation studies with test-beam data of the Geant4 simulation and Pixel.
Lloyd Algorithm K-Means Clustering. Gene Expression Susumu Ohno: whole genome duplications The expression of genes can be measured over time. Identifying.
Modification of Geant4 simulation Detailed crystal structures included reference materials real geometry.
“GRETINA/AUX Gamma Ray Tracking software” The GT tracker MSU 10/21/2012 torben lauritsen, ANL.
Community structure in graphs Santo Fortunato. More links “inside” than “outside” Graphs are “sparse” “Communities”
Lecture 18, CS5671 Multidimensional space “The Last Frontier” Optimization Expectation Exhaustive search Random sampling “Probabilistic random” sampling.
CS-ROSETTA Yang Shen et al. Presented by Jonathan Jou.
Jet reconstruction with Deterministic Annealing Davide Perrino Dipartimento di Fisica – INFN di Bari Terzo Convegno Nazionale sulla Fisica di Alice – 13/11/2007.
CS621: Artificial Intelligence
Heuristic Optimization Methods
Haim Kaplan and Uri Zwick
Subject Name: Operation Research Subject Code: 10CS661 Prepared By:Mrs
Techniques for studying correlation and covariance structure
Problem Definition Input: Output: Requirement:
Introduction to Simulated Annealing
Boltzmann Machine (BM) (§6.4)
Xin-She Yang, Nature-Inspired Optimization Algorithms, Elsevier, 2014
clustering : deterministic annealing filter (DAF)
Research algorithm of overlapping clusters in PrimEx
EMC Simulation Studies SuperB Collaboration Workshop LNF 1/12/2009
Simulated Annealing & Boltzmann Machines
Stochastic Methods.
Presentation transcript:

A new approach for the gamma tracking clustering by the deterministic annealing method François Didierjean IPHC, Strasbourg

1. Principle of the deterministic annealing method in (a few !) equations  2. Description of the algorithm  3. 2D example 4. Simulated events with AGATA detector

* set of input data (N points) goal : to assign each data point to a cluster. * set of n points, called prototypes : centers of the cluster A few definitions : * the global distortion association probability of v i with c j squared Euclidean distance

The goal of clustering is to minimize the global distortion D with the minimum of prototypes. In analogy with the statistical physics : FREE ENERGY S : Shannon entropy T : Lagrange multiplier, called now temperature The minimization of the free energy, F, consists in maximizing the entropy, S, and minimizing the distortion, D.

Deterministic annealing method : minimize F while lowering T 1. minimizing F fixing the prototype position : 2. minimizing F fixing the association probability :

As the temperature is lowered, the system undergoes a sequence of phase transitions, which consists of a natural split of the cluster. phase transition [K. Rose el al. Phys. Rev. Lett 65 (1990) 945] : A cluster centered at prototype c j undergoes a splitting phase transition when the temperature reaches the critical value T crit : 2 max. max is the largest eigenvalue of the covariance matrix :

D.A. ALGORITHM Set of 5 points (1D) V = {x i } = {0.6, 1.2, 1.5, 2.8, 3.2} initializing step : T  (all data points belong to the same cluster)  1 prototype located in the center of the data distribution

search for prototype position at constant T value convergence test : step (k)   [k] 1 -  [k-1] 1 <  T = 4  1 = 1.86

check the condition for the phase transition 2 max = 1.93 < T = 4 (critical temperature not reached) cooling step T  a T (a < 1) …… T = 1.85  1 = max = 1.93 > T = 1.85 (critical temperature reached)  splitting of the prototype

adding a new prototype position F T = 2 T = 1.86    2 [0] =  1 +   1 [0]   1 - 

1/T position Parameters of the D.A. method : * cooling factor * minimal temperature to stop the algorithm

p(c 1 /v 1 )=1. p(c 2 /v 1 )= p(c 3 /v 1 )= p(c 4 /v 1 )= p(c 1 /v 2 )= p(c 2 /v 2 )=1. p(c 3 /v 2 )= p(c 4 /v 2 )= p(c 1 /v 3 )= p(c 2 /v 3 )= p(c 3 /v 3 )=1. p(c 4 /v 3 )= p(c 1 /v 4 )= p(c 2 /v 4 )= p(c 3 /v 4 )= p(c 4 /v 4 )=1. cluster 1 cluster 2 cluster 3 cluster 4 v1v1 v2v2 v3v3 v4v4

Simulated events from AGATA : GEANT 3 code x y crystal n°19 crystal n°10 cristal n°88 crystal n°90 crystal n°148 crystal n°127 crystal n°138 simple event :  multiplicity : M  = 5 gammatypecrystalsectordeposed E 1Compton Compton Compton Photo abs Compton Compton Compton Compton Photo abs Compton Compton Compton Compton Compton Photo abs Compton Compton Compton Compton Photo abs

CONCLUSION Starting with one unique cluster, the phase transition mechanism of the deterministic annealing allows to create new cluster progressively while the temperature decreases. And, at the end of the process, the best partition is found. The first test on the deterministic annealing algorithm with 1 MeV  -ray and multiplicity 5 allows to show that the clustering method is promising.

TO DO LIST 1.Check the parameters of deterministic annealing method with different mumtiplicities using GEANT 3 : - cooling factor - minimal temperature to stop the process 2. Comparison between the 2 clustering methods with data given by the collaboration (GEANT 4) : - forward tracking / relative angle distance - deterministic annealing