Computer Scientist
Greater Boston Area
- Past
-
- Graduate Student at DEMO Lab, Brandeis University
- Student at Complex Systems Summer School, Santa Fe Institute
- Senior Applications Engineer at Oracle
- Applications Engineer at Oracle
- Programming Intern at Pretzel Logic Software Inc
- Researcher at Undergraduate Research Summer Institute, Vassar College
- Education
-
- Brandeis University
- University of Rochester
- Vassar College
- Connections
-
73
connections
- Industry
- Computer Software
- Websites
Keki Burjorjee’s Summary
In a wide range of domains, natural computational systems have abilities that far surpass those of man-made systems. It seems reasonable to suspect that there are certain core computational problems that can be solved much more efficiently than traditionally supposed, and that natural computational systems derive their impressive abilities from the extraordinary efficiency with which they solve these core computational problems. I'm interested in identifying these core efficiencies, and in using them to address computational problems that we face as a society.
Over the past few years I've focused on the simple genetic algorithm. I find it striking that this simple, biologically-plausible model of evolution routinely procures good, often great, solutions to a wide range of poorly understood combinatorial optimization problems. Despite being a highly abstract model of natural evolutionary systems, the simple genetic algorithm seems to be successful at harnessing something of the core computational efficiency that underlies the remarkable adaptive capacity of natural evolution.
I've developed a new hypothesis about the nature of this core efficiency. This account, called the generative fixation hypothesis, departs from the the reigning hypothesis of the field---the building block hypothesis---at a fundamental level. The potential impact of the generative fixation hypothesis extends beyond genetic algorithmics to (amongst others) the fields of evolutionary computation, combinatorial optimization, machine learning, and evolutionary biology.
Keki Burjorjee’s Specialties:
evolutionary computation (especially genetic algorithms), combinatorial optimization, machine learning, artificial intelligence, complex systems, computational learning, statistics
Keki Burjorjee’s Experience
-
Graduate Student
DEMO Lab, Brandeis University
(Computer Software industry)
September 2002 — August 2009 (7 years )
• Solved a hard theoretical problem in the field of evolutionary computation by finding non-trivial conditions under which the dynamics of simple genetic algorithms with infinite populations can be coarse-grained.
• Developed and submitted the Generative Fixation Hypothesis, a new explanation for the remarkable, yet mysterious adaptive capacity of simple genetic algorithms. -
Student
Complex Systems Summer School, Santa Fe Institute
(Computer Software industry)
June 2002 — July 2002 (2 months)
Attended lectures on complex systems, co-wrote and presented a survey of the literature on developmental encodings in evolutionary computation.
-
Senior Applications Engineer
Oracle
(Public Company; ORCL; Information Technology and Services industry)
August 1999 — August 2000 (1 year 1 month)
• Business Policy Framework: Invented a language for coding business policies. Wrote an interpreter for this language in Java.
• Role Based Security Framework: Implemented a security framework for the Internet Product Development module of Oracle's B2B Exchange Server. Used Java and Java Server Pages. -
Applications Engineer
Oracle
(Public Company; ORCL; Information Technology and Services industry)
August 1998 — August 1999 (1 year 1 month)
• XML Marshaller: Implemented a rudimentary object to relational mapping layer for transferring data between transient business objects and persistent XML files. Used Oracle's in-house BC4J (Business Components for Java), XML Parser, XSLT Processor and JServer tools. Provided feedback to the BC4J development team.
-
Programming Intern
Pretzel Logic Software Inc
(Computer Software industry)
June 1997 — August 1997 (3 months)
Designed and developed an installer and configuration manager in Java for a complex client-server application.
-
Researcher
Undergraduate Research Summer Institute, Vassar College
(Computer Software industry)
May 1996 — August 1996 (4 months)
Implemented a load balancing algorithm in C/C++ to make a high accuracy volume renderer run efficiently on a multi-node IBM SP2 supercomputer. Used the message passing interface (MPI) for inter-node communication.
Keki Burjorjee’s Education
-
Brandeis University
Ph.D , Computer Science , 2002 — 2009
• Dissertation title - Generative Fixation: A Unified Explanation for the Adaptive Capacity of Simple Recombinative Genetic Algorithms
• Advisor: Jordan B. Pollack, Dynamic and Evolutionary Machine Organization (DEMO) Lab -
University of Rochester
Brain and Cognitive Sciences (BCS) Department 2000 — 2002
-
Vassar College
B.A. , Computer Science (Honors); Mathematics (Honors) , 1994 — 1998
- Activities and Societies:
- Rowing Team (Captain 1996-1997)
Additional Information
Keki Burjorjee’s Websites:
Keki Burjorjee’s Interests:
evolutionary computation, genetic algorithms, machine learning, optimization, complex systems
Keki Burjorjee’s Honors:
TinyGA competition Winner, Genetic and Evolutionary Compuation Conference (2006).
Sproull Fellowship (for unusually strong potential for graduate study, University of Rochester, 2000)
Mary Evelyn Wells and Gertrude Smith Prize (for excellence in the study of undergraduate mathematics, Vassar College, 1998)
General Honors, Honors in Mathematics, Honors in Computer Science, Vassar College (1998)
Keki Burjorjee’s Contact Settings
Interested In:
- career opportunities
- new ventures
- expertise requests
- reference requests
- getting back in touch
