ATTİLA AŞKAR 70th BIRTHDAY CONFERENCE

23 May 2014

Koç University

Speakers

Mehmet Can	Int. Univ. of Sarajevo
Burak Erman	Koç University
Gerard Maugin	Paris 6
Herschel Rabitz	Princeton University
Alphan Sennaroğlu	Koç University
Mete Soner	ETH - Zurich

Conference Photos

Please click the slideshow link in the upper left corner.

conference poster

Program

9:00	Opening Remarks
9:30	Herschel Rabitz	Control of Quantum Phenomena in Physics and Chemistry
10:30	Coffee Break
10:50	Gerard Maugin	Generalized continuum mechanics
11:50	Break
12:00	Mete Soner	From Thales and Gaspard Monge to VIX options
13:00	Lunch
14:30	Burak Erman	Statistical Thermodynamics of Proteins
15:30	Coffee Break
15:50	Mehmet Can	Human Chromosome Classification
16:50	Break
17:00	Alphan Sennaroğlu	Graphene and carbon nanotube saturable absorbers
18:00	Closing Remarks

Abstracts

Herschel Rabitz

Control of Quantum Phenomena in Physics and Chemistry: Learning from Each Other

Control in the physical, chemical, and biological sciences is pervasive. The dynamics involved span vast length and time scales with the associated controls ranging from shaped laser pulses out to the utilization of chemical reagents and processing conditions. Despite all of these differences, there is common behavior found upon seeking optimal control in these various domains. Evidence for this common behavior will be presented with the most manifest finding that efforts to seek an optimal control are far more efficient than the apparent complexity of the applications would suggest. Illustrations will be provided for control of quantum phenomena and in the chemical sciences in order to assess the common principles and illustrate the complimentary insights they provide. The potential significance of the findings will be discussed.

Gerard Maugin

 Generalized continuum mechanics with applications to biomechanics

Although sometimes referred to original works by Galileo Galilei in solids, John Bernoulli for the mechanics of muscles, and Poiseuille for the flow of fluids, it is only recently that biomechanics succeeded to exploit efficiently modern tools of continuum thermo-mechanics including some advanced aspects such as microstructure in generalized continuum mechanics, mutiplicative decomposition of finite deformation, theory of mixtures, notion of internal variables of thermodynamic state, higher-order gradient theory, nonlinear waves, configurational forces, and non-Riemannian material space. These advances are illustrated in non-technical terms by recent works of the author and co-workers, essentially with problems posed by the growth of long bones in mammals and the simulation of the growth of some biological tissues such as in the expansion of the horn of some animals and the developments of some sea shells. This is an occasion to pay a tribute to our “Turkish connection” which started with Prof. Ratip Berker in Paris, Prof. A.C. Eringen in Princeton and, obviously, continued for so many years with Attila Askar in Princeton, Istanbul and Paris.

Mete Soner

From Thales and Gaspard Monge to VIX options

Modern financial industry makes use of advanced mathematical techniques in its daily operations. They are mainly used for risk quantification and management. Many new complex financial instruments are created through these tools. In turn they provide the needed additional financing for many vital industries. On the scientific side, mathematical finance provides an excellent laboratory for the general question of decision making under uncertainty. In this survey talk, I will outline the historical developments and discuss few current applications.

Burak Erman

Statistical Thermodynamics of Proteins

Proteins perform their function through organized fluctuations of their atoms. Evolutionary processes, allostery, catalytic activity all depend on the correlated fluctuations of the amino acids of proteins. In this talk, an overview of the statistical thermodynamics basis of energy and residue position fluctuations will be given and relations between protein architecture and function will be discussed. It will be shown that correlated fluctuations may be manipulated by binding drug molecules at suitable locations on the protein. Examples of change of function by binding drug molecules will be presented.

Mehmet Can

Human Chromosome Classification Using Competing Neural Network Teams Supplemented by Mixed Signals Information (CNNT-S)

This talk presents a new approach to human chromosome classification. Human cell contains 22 pairs of autosomes and a pair of sex chromosomes. In this research, 22 types of autosomes represent 22 classes to be distinguished. New method of classification is based on the special organized committee of 462 simple perceptrons P(i,j), called Competitive Neural Network Teams, (CNNT)'s. Each perceptron P(i,j) is trained to differentiate type i chromosomes, from type j chromosomes, hence there are 22 x 21 learning machines. Moreover, for completeness, dummy perceptrons P(i,i) are added to the team which flags always zero. The final outcome of the testing data is a 22x22 decision matrix, containing outcomes of each of 22x22 perceptrons in the team. When a chromosome of type k is presented to the 22x22 committee of perceptrons, in general perceptrons of the column P(i,k),i=1,2,…,22;i≠k flags 1, while the perceptrons of the row P(k,j),j=1,2,...,22;j≠k flags -1. The score of a team is the sum of the number of 1's in the kth column and the number of -1's in the kth row.
The column-row pairs {P(i,k),P(k,j) },k=1,2,…,22 are called competing teams of Neural networks. When a chromosome of type m is presented to the 22 teams of Neural networks, the team m is expected to have the highest score, and be the winner of the competition. We identify the presented chromosome's type with the label of the winning team. This technique yields a significantly high correct classification rate of 95%.
For a new chromosome data, if the perceptron P(i,j) flags 1, but the perceptron P(j,i) flags also 1, or vice versa, this case is called a mixed signal case, and the chromosome is put into the “indecisive cases” basket. It is seen that the method (CNNT) can be further improved by reclassification of the chromosomes in the “indecisive cases” basket by the use of Nearest Neighbor technique. This classification technique which we call CNNT-S is applied to the Copenhagen chromosome dataset and correct classification rate is improved over 97.27%.

Alphan Sennaroğlu

Graphene and carbon nanotube saturable absorbers for femtosecond pulse generation from solid-state lasers

The presentation will review our recent saturable absorber experiments which were initially inspired by a collaborative theoretical investigation conducted with Professor Aşkar in mid 1990s. Saturable absorbers can be used to initiate pulsed operation of solid-state lasers, in which optical gain is achieved by using insulating crystals, glasses, or ceramics doped with laser-active ions. Due to their broad tuning range, vibronic solid-state lasers, such as Cr:ZnSe, are also suitable for the generation of femtosecond pulses. The first part of the talk will focus on the general physical characteristics of solid-state lasers, some of their applications, and different regimes of operation. In the second part, I will discuss some of the emerging methods of femtosecond pulse generation, employing graphene and carbon nanotube devices as saturable absorbers. The presentation will further describe our recent experiments, conducted in Koç University Laser Research Laboratory, in which carbon nanotube and graphene saturable absorbers were used to generate femtosecond pulses from vibronic solid-state lasers operating near the wavelengths of 1.3 um (near infrared) and 2.4 um (mid infrared).

The conference is supported by Koç University.

For more information, please send an e-mail to askarconference@gmail.com

Transportation to Koç University

There will be shuttles from Haciosman metro station to Koc University at 8:45 and 10:15. The return shuttle will be at 18:15.