Computer Science Engineering

Education

PhD ****~presen

Computer Science (Artificial Intelligence and Machine t

Learning), University of Lugano (Computer Science department

is ranked 3rd in Switzerland) & University of Massachusetts

at Amherst (The artificial intelligence program is ranked

8th in the U.S)

Research projects:

Modeling Housing Market Dynamics Using a Multi-Agent

Simulation of Participants' Cognitive Behavior.

The housing market is modeled as an adaptive complex system

using the multi-agent based modeling approach. By modeling

the decision-making and reasoning of the agents who

participate in the housing market, study will be made of the

dynamics and evolution of housing markets in Lugano and

surrounding towns, and the related segregation effects due

to different ethnic compositions of neighborhoods.

Generating Trading Rules Using Nash-Genetic Programming

Genetic-Nash algorithm is used to combine genetic algorithm

and Nash strategy in order to cause the genetic algorithm to

build the Nash equilibrium. In this research, Genetic-Nash

Programming (GNP), inspired by Genetic-Nash algorithm and

genetic programming, has been introduced and developed.

Nash-Genetic programming is exploited to generate trading

rules that are devised to generate appropriate buying and

selling signal over short time periods. Here we have used

historical pricing and transaction volume data in order to

generate proper decision about selling or buying share in

each day.

M.Sc. 2004~2006

Computer Science (Artificial Intelligence and Machine

Learning), Amirkabir University of Technology (Polytechnic

of Tehran), Tehran, Iran

B.Sc. 2000~2004

Applied Mathematics, Iran University of Science and

Technology, Tehran, Iran

Profile

. Highly motivated and results-oriented PhD in computer science (Machine

Learning and AI) with fellowships and awards throughout education

. Proficiency in programming in MATLAB, C++, C, Python, R, PASCAL, SQL and

others

. Excellent quantitative/analytical skills and an ability to apply

skills/knowledge in business processes

. Quick learner, Self-starter, Excellent communication skills, Proactive,

Team player.

. A solid understanding of financial mathematics and ability to creatively

solve problems.

. Solid financial knowledge from self-study and taking some courses

(financial engineering in Isenberg School of Management: Department

of Finance and financial mathematics in Mathematics Department in Umass

Amherst)

. Seek positions in quantitative modeling

Skills

Finance Good understanding of the Principals of financial engineering:

Risk neutral valuation, Black-Scholes, Martingale, Hedging,

Greeks, and Delta hedging etc.

Algorithm Forward-Backward Algorithm, Simulated Annealing, Heuristic

Skills Local Search, Optimization Techniques, Genetic algorithm,

Genetic programming.

Modeling Partial Differential Equations, Monte Carlo, Time-Series

Skills: Analysis, Stochastic Process, Regression Techniques, Hidden

Markov Model, Neural Networks, Fuzzy Systems, Graph Theory,

Game Theory, Inference in Graphical models, Structured

probabilistic models, Bayesian Inference.

Programmin C/C++, Matlab (basis during PhD research, includes Simulink and

g various toolboxes), Python.

Database SQL

Operating Windows, OS X, Linux

system

Quantitative skills through self-study

Thorough understanding on Stock and Bond Valuation, Capital Asset

Pricing Model (CAPM), Options and Derivatives, Portfolio management,

Investment strategies, Asset allocation, Option pricing, portfolio

optimization and construction, VaR models.

Awards and Honors

Received overseas researcher scholarship award from Swiss 2010

National Science Foundation

Full Scholarships from Swiss National Science Foundation (SNSF) 2007

for three years.

Iranian State Full Scholarship for master of science studies 2004

Ranked 23rd in the Iranian nationwide M.Sc. university entrance 2004

exam in Artificial Intelligence among more than 10000

participants

Iranian State Full Scholarship for bachelor 2000

Research Experience

Research Assistant, MAS -Lab, University of Massachusetts at 2010

Amherst, MA.

Complex negotiation models in electronic commerce

Using Fuzzy Control theory and soft computing techniques, we

have designed and developed an automated negotiation model.

Research Assistant, MACS-Lab (Modeling and Application of 2009~201

Complex System Laboratory) 0

Using Nash-Genetic Programming in order to Extract Trading

Rules

Genetic-Nash algorithm is combination of Genetic algorithm and

Nash strategy in order to reach to Nash equilibrium point. In

this work, inspired by Genetic-Nash algorithm and Genetic

programming, Genetic-Nash programming has been introduced and

developed. This method could be applied in any type of market

to extract trading rules. In this work we have used this method

in order to extract housing market trading rules.

Research Assistant, IDSIA (Dalle Molle Institute for Artificial 2007~201

Intelligence), 0

Mas Lab (Multi-Agent Systems Laboratory), Switzerland

Effects of Neighborhood Choice on Housing Markets: a model

based on the interaction between micro simulations and

revealed/stated preference modeling.

For modeling housing market as a significant indicator of

economic situation in a country we have used an agent-based

modeling method and started to design intelligent agents who

are able to make interactions with other agents and the

environment and make decisions considering their behavior and

situation in the world. Our model does not rely on the

assumption that the economy will move towards a predetermined

equilibrium state; instead it has this ability that at any

given time, each agent acts according to its current situation,

the state of the world around it and the rules governing its

behavior. The challenges in this work were modeling of human

reasoning in complex environment and processing of uncertain

data. In this work, Demand and Supply will be emerged as the

outcome of the underlying complex system (housing market).

Through this work type-2 fuzzy toolbox has been developed using

Matlab.

Artificial Intelligence Lab, University of Tehran, Tehran, Iran 2006~200

7

Using of Dynamic Synapse Neural Networks (DSNN) for Noisy

Signals Processing.

This project has been developed using Matlab and Neural Network

toolbox. Signal processing has been done by wavelet

decomposition. In this study we have applied a Genetic

algorithm (GA) learning method with different fitness functions

to optimize the neural network.

Pattern Recognition Lab, Amirkabir University of Technology 2005~200

(Polytechnic of Tehran), Tehran, Iran 7

Variant Combination of Multiple Classifiers Methods for

Classifying the EEG Signals in Brain-Computer Interface

Using different methods in Signal processing and pattern

classification we have designed a Brain-Computer Interface

System in order to recognize the decision of the Brain to

either move to right or left. The result of the work was

superior in compare to the best result of the BCI Competition

in 2003.

Publications

[1] Esmaeili. M "Creating Divers classification Systems in Processing of

EEG Singnals in Human-Computer Interfaces", Twenty-Second Conference on

Artificial Intelligence (AAAI-07), Vancouver, British Columbia, Hyatt

Regency Vancouver, July 22-6

[2] Esmaeili. M, Rahmati. M, "Designing of Multiple Classifier Systems by

Fuzzy Decision Making", IEEE International Conference on Fuzzy Systems,

Imperial College,London, UK.

[3] Maryam Esmaeili, Mohamad H. Jabalameli, Zeinab Moghadam: A New Scheme

of EEG Signals Processing in Brain-Computer Interface Systems. The 2007

IEEE International Conference on Granular Computing GrC 2007: 522-527.

[4] Esmaeili M. Mogadam Z, "Channel Selection in Brain Interface Systems",

IEEE International Conference on Intelligent Systems IS'08, Bulgaria.

[5] Esmaeili. M, Shoaie. Z, Bagheri. S, "Combination Of Multiple

Classifiers With Fuzzy Integral Method for Classifying The EEG Signals in

Brain-Computer Interface", The International Conference on Biomedical and

Pharmaceutical Engineering 2006

(ICBPE2006).

[6] Emaeili. M, Rahmati. M, "A New Scheme for Feature Selection in Ensemble

with Majority Vote Combiner for EEG Signal Processing in Brain-Computer

Interface", the 13th Iranian Conference on Biomedical Engineering, 2006.

[7] Esmaeili M. Rahmati M "Using of multiple classifier systems in EEG

Signals Processing in Brain-Computer Interface", 12th International CSI

Computer Conference (CSICC2007), Tehran, Iran.

[8] Esmaeili M. Rahmati M "Bagging and Boosting Approach for EEG Signals

Classifying in Brain-Computer Interface Systems", 2007 ICEE Iranian

Conference on Electrical Engineering (ICEE 2007), Iran Telecom Research

Center, Tehran, Iran

[9] Esmaeili M. "Modeling of intelligent agent behaviors in dynamic

system", Book Chapter. In Advances in Cognitive Systems, Herts, UK:IET

Publisher.

[10] Maryam Esmaeili, Prakash L. Abad, Mir-Bahador Aryanezhad: Seller-Buyer

Relationship when End Demand is Sensitive to Price and Promotion. APJOR

26(5): 605-621 (2009).

[11] Maryam Esmaeili, Mir-Bahador Aryanezhad, Panlop Zeephongsekul: A game

theory approach in seller-buyer supply chain.European Journal of

Operational Research 195(2): 442-448 (2009)

[12] Esmaeili, M, A Vancheri, and P Giordano, Mathematical and

Computational Modeling of Housing Market Dynamics - System engineering

point of view." IEEE International Systems Conference 2010. San Diego, CA,

2010.

[13] Esmaeili, M, A Vancheri, and P Giordano. "Extracting the Trading Rules

in Housing Market Using Nash Genetic Programming Approach." 16th

International Conference on Computing in Economics and Finance. London,

2010.

[14] Esmaeili, M, A Vancheri, and P Giordano, Modeling of Demand in Housing

Market through Multi-Agents Behavioral Modeling, ECCS'10 European

Conference on Complex Systems. Lisbon, 2010.

[15] Esmaeili, M, A Vancheri, and P Giordano, Modeling of housing market

as an adaptive complex system. 22nd conference of the European Network for

Housing Research, Urban Dynamics and Housing Change. Istanbul, 2010.

[16] Esmaeili, M, and A Vancheri. "A new Approach in Cooperative Decision

Making in Multi-Agent System Inspired by Human Visual Cortex." 2010 IEEE /

WIC / ACM International Conferences in Intelligent Agents Technology.

Toronto, 2010.

Review

1- COGSCI 2009 The annual meeting of the cognitive science society

2- IEEE Transactions on Knowledge and Data Engineering (TKDE)

3- 2007 ICEE Iranian Conference on Electrical Engineering (ICEE 2007),

Iran Telecom Research Center

4- The First International Conference on Complex Sciences: Theory and

Applications

Contact this candidate