1. Curriculum Vitae
Software engineer (Algorithm), Physicist, Statistician
14/8 Nahal Soreq street,
Modi'in, Israel
Tel: 08-9705640, 052-2813942
E-mail: gleshem@cc.huji.ac.il
Url: http://shum.huji.ac.il/~gleshem
Personal data:
Full Name: Guy Leshem
Place of Birth: Kibbutz Ein Gav, Israel.
Birthdate: Aug-24-1963
Nationality: Israeli, German.
Marital status: Married to Shirely, Father to Shani.
Pets: Snow (Dog).
Address for Correspondence:
Department of Statistics
The Hebrew University of Jerusalem
Mount Scopus, Jerusalem 91905
ISRAEL
Highlights:
• PhD thesis that dealing with a new algorithm (development) in Statistics Machine
learning (based on Adaboost and Random Forest), Machine Vision (based on
Traffic Surveillance) and Operational Research (based on "Queuing systems
consisting of N queues served by a single server"). Obtained in Nov 2002 from
Hebrew university of Jerusalem.
• MS.c. laboratory work on: "Computer Add Surgery Development new algorithm
for reconstruction of 3D shapes views and statistical model".
• Ten years working as a networking engineer for Intel on the development of a
new GUI for networking configuration, Advance Network Server (ANS), SNMP
drivers on Linux OS.
• Statistical Adviser (Working at Applied Statistics Laboratory) for number of
commercial companies and academic researcher.
• Wide-ranging experience in tutoring and teaching with a good record of
motivating, challenging and supporting students.
• Creative, versatile and communicative.
• Two languages spoken fluently.
• Battalion Commander in army reserve (degree Lieutenant Colonel).
2. Professional Experience:
Work Experience:
• 10 years work experience in Intel as Software Engineer and Process
Engineer:
• Software Engineer in Intel IDC (networking Group), dealing with the
development and testing of adapters drivers for Linux / UNIX Os. The
projects that I deal as following:
1. Development of a new GUI for networking configuration of Linux
server for Dell Corporation. This GUI (PROCfg) is a reporting and
configuration tool for Intel adapters and Intel Advanced Network
Services (iANS). It works with the Intel drivers and running Linux.
2. Development of Advance Network Server (ANS) (teaming of number
of adapters) for UNIX Os (SCO). This tool support the following
teaming modes: Adapter Fault Tolerance, Switch Fault Tolerance,
Adaptive Load Balancing etc.
3. Development SNMP drivers on Linux Os for communication between
Linux server and windows client with MG-Soft. This SNMP agent
extension extends the UCD-SNMP agent for Linux the SNMP
protocol. It provides information on the Intel the adapter drivers,
teaming (Advanced Network Services (ANS) and VLANs,).
On a human and organizational level, the experience at
Intel gave me a first-hand impression of various leadership
styles and of the difficulties that can occur when managing
technical development projects.
• Process Engineer in Intel, dealing with development of a new process for
chips manufacture, control on the existing process and analysis of end of
line data (E-test, Sort), 1993 – 1999.
• Statistical Adviser
• Working at "Applied Statistics Laboratory" on numbers of project for
commercial companies and academic researcher. At this lab I’m working
on research project of MA/PhD students from the design to the final
analysis under the supervision of Prof. Ya'acov Ritov, 2004-2005, and on
projects for commercial companies under the supervision of Dr. Ronit
Nirel (2006-2007)
• Teaching Experience:
Courses taught:
Introduction to Statistics (for non-statisticians) (2006).
PROGRAMMING FOR STATISTICAL APPLICATONS (2007).
3. Other:
Teaching Assistant for Professor Leo Joskowicz, Robotics II, 1998.
Teaching Assistant for Dr. E. Zuchbistky, Matrices and Vector, 2002.
Teaching Assistant for Professor David Zucker, Regression and analysis, 2002
• Academic Experience:
1. Doctorate research, My PhD study include development a new algorithm for
improving the accuracy of algorithms for learning binary concepts. The
improvement is achieved by combining Random forests algorithm into AdaBoost
as weak learner. Random forests are a combination of tree predictors, where each
tree in the forest depends on the value of some random vector ,טso this weak
learner expect to be more “strong” and more accuracy, since 2003.
Additionally, my Ph.D. thesis also include the development of two new algorithm,
the first deal with real-time image processing of video stream for car tracking in
traffic video and the second with Operational Research model "Queueing systems
consisting of N queues served by a single server".
2. Master degree at the Computer Science institute Hebrew University. Work on
Image protocol and developing of Generic hardware, algorithms, and protocols
that will allow the accurate registration as a Lab study, 1999 – 2003. The project
that I deal with was developing innovative computer-based methods for assisting
surgeons in the planning, execution, and evaluation of surgical procedures based
on medical images.
3. Master Work "Developing of a Room Temperature KLAN Pyroelectric Detectors
Array for Thermal Imaging at the Graduate school of applied science &
Technology", 1993 – 1996. The project that I deal with was developing new
thermal imaging technology based on room temperature KLAN Pyroelectric
detectors array. The principal technical features were (1) Long-wavelength IR
operation. (2) Room temperature KLAN Pyroelectric detectors array. (3) Applied
electronic field to boost responsivity and provide stability. (4) Use of mechanical
chopper for field-difference processing. (5) Bump bonding of detector to readout
the electric signal.
Education:
2003 - 2007 : Ph.D. study at the Statistic department in Hebrew university of Jerusalem.
The Ph.D. subject: "Improvement of Adaboost Algorithm by using Random Forests as
Weak Learner and using this algorithm as statistics machine learning for traffic flow
prediction" under the supervision of Prof. Ya'acov Ritov.
• Thesis Topic: Advance technologies in identification and prediction of urban
traffic for the prevention of traffic violation and accidents (Improvement of
Adaboost Algorithm by using Random Forests as Weak Learner and using this
algorithm as statistics machine learning for traffic flow prediction).
4. • Principal Thesis Advisors: Prof. Yaacov Ritov.
• Major Fields of Interest: Development algorithm for Statistics Machine
learning, Machine Vision Based Traffic Surveillance and Operational Research
model "Queueing systems consisting of N queues served by a single server".
1999-2003: M.Sc. study in Hebrew university of Jerusalem at the Computer Science
department, part of the time guidance by Prof. Leo Juscovitch. The M.Sc. study includes
laboratory work on: "Computer Add Surgery Development new algorithm for
reconstruction of 3D shapes views and statistical model". (Mark: 85)
1993-1996: M.Sc. study in the Hebrew university of Jerusalem at the Graduate school of
applied science & Technology, guidance By Prof. A.J.Agranat. The M.Sc. subject:
"Development of a Room Temperature KLAN Pyroelectric Detectors Array for Thermal
Imaging". (Mark: 85)
1988-1992: B.Sc. study "material engineering for micro - electronic" in Ben Gurion
university at the Material Engineering department. (Mark: 84)
Publications:
• "Traffic Flow Prediction using Adaboost Algorithm with Random Forests as a Weak
Learner" Guy Leshem, Ya'acov Ritov. XIX. International Conference on Computer,
Information, and Systems Science, and Engineering January 29-31, 2007 Bangkok,
Thailand. [pdf (498)].
• "FRACAS: A System for Computer-Aided Image-Guided Long Bone Fracture
Surgery" L. Joskowicz, C. Milgrom, A. Simkin, O. Sadowsky, Z. Yaniv, G.Leshem
Proc. of the 2nd Int. Symposium on Medical Robotics and Computer Assisted
Surgery ,Jeruslem, 1999.
• "In-vitro accuracy of contact-based registration: materials, methods, and experimental
results", Z. Yaniv, O. Sadowsky, G. Leshem, L. Joskowicz, Technical report,
December 1999. [pdf (568)].
• Computer-aided image-guided bone fracture surgery: system integration and
prototype" L. Joskowicz, C. Milgrom, A. Simkin, O. Sadowsky, Z. Yaniv and G.
Leshem. Proc. 13th Int. Symposium on Computer Assisted Radiology and Surgery,
H.U. Lemke et. al. eds, Elsevier 1999, pp 710 [pdf (0.06MB)].
Major Practice:
.++Image processing by C•
.Robot movement analysis by Matlab•
• Development of algorithms for statistics machine learning: Adaboost and Random
Forests (written in Matlab, Fortran and C++).
• Development of algorithms for statistics analysis of missing data (written in
SAS).
5. • Compilers course by Java.
Skills:
• Scientific:
1. Due to Ph.D. in statistic and two master degrees (Physics & Computer
science) I have excellent ability to deal with a new issue.
2. Excellent knowledge of in Statistics Machine learning Theory, Machine
Vision and Operational Research model.
3. Good base in development of Adapters Drivers and Computer Networks.
• Programming: Very experienced Visual Studio C++, C, SAS, SQL, Matlab,
Windows, Unix and Linux.
Some experience of JAVA. SAS, R, SQL.
• Languages: Hebrew (mother tongue), English (high level).
Army:
Parachutist special unit, I'm battalion commander in army reserve (degree Lieutenant
Colonel).