Panos Tsilifis Agia Anna

Panagiotis Tsilifis


Postdoctoral Scholar / Research Associate



3620 South Vermont Ave. KAP 210,
Los Angeles, 90089-2532, CA, USA
email: tsilifis@usc.edu

tel: +1 (213) 740-9165



Research Interests

Uncertainty Quantification : Bayesian inverse problems, Polynomial Chaos, uncertainty propagation
Statistics : Bayesian statistics, design of experiments, Monte Carlo methods
Nonlinear dynamics : Soliton propagation, Bose-Einstein condensates, numerical simulations

Education

Ph.D. in Applied Mathematics, USC - 08/2016
Thesis : Design, Adaptation and Variational Methods in Uncertainty Quantification

M.A. in Applied Mathematics, USC - 08/2014
M.Sc. in Applied Mathematics, NTU Athens, Greece - 06/2011
Thesis : Nonlocal Soliton Dynamics (pdf in English... ...and in Greek)

Erasmus visiting student KTH Stockholm, Sweden - 08/2008-06/2009
Diploma in Applied Mathematical and Physical Sciences, NTU Athens, Greece - 09/2009
Thesis : Stability of Solitons to the Nonlinear Schrödinger equation
Agia Anna


Research work - Publications

Journals
· The stochastic quasi-chemical model for bacterial growth: Variational Bayesian parameter update
P. Tsilifis, W.J. Browning, T.E. Wood, P.K. Newton and R.G. Ghanem
Journal of Nonlinear Science (2017)
· Homogeneous Chaos basis adaptation for design optimization under uncertainty: Application to the oil well placement problem
C. Thimmisetty, P. Tsilifis and R.G. Ghanem
Artificial Intelligence for Engineering Design, Analysis and Manufacturing 31 (3) 265-276 (2017)
· Reduced Wiener Chaos representation of random fields via basis adaptation and projection
P. Tsilifis and R.G. Ghanem
Journal of Computational Physics 341 102-120 (2017)
· Efficient Bayesian experimentation using an expected information gain lower bound
P. Tsilifis, R.G. Ghanem and P. Hajali
SIAM/ASA Journal on Uncertainty Quantification 5(1) 30-62 (2017)
· Computationally efficient variational approximations for Bayesian inverse problems
P. Tsilifis, I. Bilionis, I. Katsounaros and N. Zabaras
ASME Journal of Verification, Validation and Uncertainty Quantification, 1(3) 031004 (2016)
· Markov Chain Monte Carlo inference of parametric dictionaries for sparse Bayesian approximations
T. Chaspari, A. Tsiartas, P. Tsilifis and S. Narayanan
IEEE Transactions on Signal Processing, 64(12) 3077 - 3092 (2016)
· Cubic-quintic long-range interactions with doule-well potentials
P.A. Tsilifis, P.G. Kevrekidis and V.M. Rothos
Journal of Physics A, 47(3) 035201 (2014)

Conferences - Presentations
Proceedings :
· Variational reformulation of Bayesian inverse problems
P. Tsilifis, I. Bilionis, I. Katsounaros and N. Zabaras
ECCOMAS 1st International Conference on Uncertainty Quantification in Computational Sciences and Engineering (UNCECOMP2015), Crete island, Greece, 25-27 May 2015
Invited Talks :
· Efficient methods for Bayesian experimental design and inference and dimensionality reduction for Polynomial Chaos expansions
Combustion Research Facility, Sandia National Labs, Livermore, CA, USA, September 27, 2017
· Gaussian Process regression for Sovereign Ratings forecasting
MIGA Department, World Bank Group, Washington, D.C., USA, November 15, 2016
Talks/Abstracts :
· Variance reduction methods for efficient Bayesian experimental design
P. Tsilifis, R. Ghanem
SIAM Conference on Uncertainty Quantification (UQ16), EPFL Campus, Lausanne, Switzerland, 5-8 April 2016
· Variational Bayesian inference using basis adaptation in Homogeneous Chaos surrogates
P. Tsilifis, R. Ghanem
SIAM Conference on Uncertainty Quantification (UQ16), EPFL Campus, Lausanne, Switzerland, 5-8 April 2016
Posters :
· Optimal Bayesian experimental design for subsurface characterization of contaminated areas
P. Tsilifis, R. Ghanem, P. Hajali
CNLS Conference on Data Analysis (CoDA2016), Santa Fe, NM, USA, 2-4 March 2016
· Markov Chain Monte Carlo inference of parametric dictionaries for sparse Bayesian approximations
T. Chaspari, A. Tsiartas, P. Tsilifis, S.S. Narayanan
2015 MBMC Workshop: Communications, Inference and Computing in Molecular and Biological Systems, USC, Los Angeles, CA, USA, 3 & 4 December 2015
· Optimal Bayesian experimental design for contaminant transport under uncertainty
P. Tsilifis, R. Ghanem
USACM 13th US National Congress on Computational Mechanics (USNCCM13), San Diego, CA, USA, 26-30 July 2015
· Optimal Bayesian experimental design for permeability identification in the presence of contaminants
P. Tsilifis, R. Ghanem
SIAM Conference on Mathematical and Computational Issues in the Geosciences (GS15), Stanford University, Stanford, CA, USA, 29 June-2 July 2015
· Variational reformulation of Bayesian inverse problems
P. Tsilifis, I. Bilionis and N. Zabaras
SIAM Conference in Computational Sciences and Engineering (CSE15), Salt Lake City, UT, USA, 13-18 March 2015

Graduate Coursework

USC :
Real Analysis, Numerical Analysis, Introduction to Mathematical Statistics I and II, Probability Theory I and II, Econometric Methods, Economic and Financial Time Series, Analysis of Variance and Design, Stochastic Differential Equations, Stochastic Simulation, Filtering Theory, Uncertainty Quantification, Computational Reservoir Modeling, Combinatorial Analysis, Applied Matrix Analysis, Chaos Expansions and Malliavin Calculus.
NTUA :
Functional Analysis, Numerical Analysis and Finite Elements, Dynamical Systems and Chaos theory, Time-frequency Analysis and Wavelets, Nonlinear Functional Analysis, Harmonic Analysis, Partial Differential Equations, Nonlinear PDEs and Variational Methods.


CV

Teaching

Fall 2011
Spring 2012
Fall 2012
Spring 2013

Links

View Panagiotis Tsilifis's profile on LinkedIn