| Klimasauskas Group staff have a
wide range of capabilities. This section summarizes them by "topic".
One of the value aspects of our wide-ranging experience is that we are
able to create innovative solutions to new problems.
Business Applications
- Value propositions based on intermediate key performance indicators
for both cost justification and post-project performance audit.
- Proforma analysis of complex projects complete with "what-if"
capabilities for quantifying risk profiles.
- Product development management.
- Knowledge Management, primarily focused on driving innovation.
- See also: Financial Applications.
Training Courses Developed / Taught
Technology Applications
- Work-process integration and Neural Network
automation (Klimasauskas Group client)
- Dynamic on-line demographic profiling (Klimasauskas Group client)
- Monte Carlo simulation for estimating risk of ruin (Advanced Technology
for Developers)
- Trading signal based on optimized
mixture of MACD oscillators (Advanced Technology for Developers)
- Genetic algorithm for identification of neural network architecture
and weights for artificial trading signal (NeuralWare)
- Genetic algorithm for identification of
state-space parameters for inferential model based on laboratory samples
(AspenTech)
- Hybrid model for modeling non-linear physical
processes (NeuralWare)
- Plasma etch maintenance advisor
(NeuralWare / Digital Equipment Corporation)
- Genetic algorithms applied to variable selection (NeuralWare / DARPA)
- Automated data analysis and transformation (NeuralWare)
- Data conflict reduction by Rubic's Cube (NeuralWare)
- Data sub-selection algorithms for improved model performance on "lumpy"
data sets (NeuralWare)
- Automated stock portfolio management
(NeuralWare client)
- Process fault identification using fuzzy encoding of continuous inputs
(NeuralWare client)
- True name fraud detection (NeuralWare
client)
- Credit card fraud detection (various NeuralWare clients)
- Name selection for direct mail campaigns
(various NeuralWare clients)
- Parallel (Intel Hypercube) computing port of NeuralWorks Professional
II/Plus (NeuralWare / ParaSoft)
- NeuralWorks Professional II/Plus design, implemented 18 major paradigms
(NeuralWare)
- Schematic back-annotation system (Data IO)
- Improved and debugged routing algorithms for automatic printed circuit
board layout package (Data IO)
- High-speed graphic interface for printed circuit board layout package
(Data IO)
- Proto-type part inspection system using
IRI SIMD vision system and Visual Machine's Limited algorithms (American
Industrial Vision)
- Proto-type part location and tracking system using IRI SIMD vision
system (American Industrial Vision)
- Real-time adaptive welding proto-type
using Meta Machines Limited robotic welding head (American Industrial
Vision)
- Teach-pendant based language for controlling the Merlin Robot (American
Robot)
- Real-time operating system for robotic control, including real-time
kernal, file system, and i/o (American Robot)
- High Density Optical Disk - key member of design team for control
firmware (Xerox EOS)
- High speed visual inspection system
- design team for image pre-processor, implemented real-time analysis
and classification component. (Xerox EOS)
- Kontron In-circuit emulator diagnostics package (Kontron Electronics)
- Universal multi-machine linkage editor for in-circuit development
system (Future Data)
- 1820 In-circuit emulator controller (Future Data)
- Failure Data Analysis system for semiconductor tester (Zehntel)
- High reliability fault-tolerant file system. (Future Data)
- FORTH cross-compiler (Zehntel)
- 6800 Basic compiler (Future Data)
- Universal linkage editor design and implementation. (Future Data)
- Financial Analysis: Income property, multi-use real estate development,
charitable remainder trust proforma. (GOCO)
- Tagged data-base design (similar to XML) for storing sparse ethno-demographic
data (1977) (World Vision)
- Fortran-callable un-buffered random access variable length file system
in support of simulation of a Liquid Metal Fast Breeder Reactor (LMFBR)
simulator (Argonne National Laboratory)
- Automotive Fuel Efficiency and Maintenance study (Department of Transportation)
- Brute-force search program for seeking out high-performance linear
9x3 convolutional codes (Jet Propulsion Laboratory)
- Real-time analysis system for measuring instruction mix (Jet Propulsion
Laboratory)
Technical Capabilities
- Neural Networks - implemented 18 major network paradigms, and many
variants. Key designer of NeuralWare's
NeuralWorks Professional II/Plus, and NeuralWorks Predict products.
Developed and taught courses on Neural Networks in Business, Industry,
and Government; Market Timing with Neural Networks; Neural
Networks in Process Control. See also: NeuralWare
consulting.
- Genetic Algorithms - implemented genetic optimization using bit-chromosomes,
integer and real-valued chromosomes, mixture-based chromosomes, and
sequence-based chromosomes. Applied to a variety of optimization applications
including route planning and scheduling, maximizing internal rate of
return subject to maximum draw-downs (multi-objective optimization),
portfolio balancing, variable selection, real-valued function optimization,
market timing, hybrid neural-genetic models. For an example of this,
see the Small Data Set Classifier developed
for DARPA.
- Fuzzy Logic - developed as a method for pre-processing data for neural
networks, fault diagnosis, non-linear process modeling.
- Dynamic Hill Climbing - applied to semiconductor maintenance advisor.
- Monte Carlo Simulation - risk of ruin, estimating probability that
a model made predictions better than random, Liquid Metal Fast Breeder
Reactor simulation.
- Image Processing - Hough transforms, Canney edge detectors, patch
and whole image analysis, FFT signature analysis.
- C/C++, Visual Basic, Java, FORTH, FORTRAN, Assembly Code.
- Microsoft Windows (95, 98, ME, 2000, NT, XP),
- Unix (Solaris, HP-UX, AIX, Linux)
Selected Publications
- Below are some selected publications not referenced elsewhere in this
web site.
- Klimasauskas, C. (1996). Fuzziness. PC
AI. May / June 1996.
- Klimasauskas, C. (1996). Neural Networks for Database Applications.
Fuzzy
Logic and Neural Network Handbook, Chen, C.H. (ed.). Chapter
11. pp. 11.1-11.12. New York: McGraw-Hill.
- Klimasauskas, C. (1993). Applying Neural Networks. Neural
Networks, Trippi, Robert R. (ed.); Turban, Efraim (ed.). Chapter
3. pp. 47-72. Chicago: Probus Publishing Company.
- Klimasauskas, C. (1993). Back-propagation with Less Information.
Advanced Technology for
Developers. September, 1993. High-Tech Communications: Sewickley,
PA.
- Klimasauskas, C. (1993). Genetic Optimization in Excel: Part II.
Advanced Technology for
Developers. July, 1993. High-Tech Communications: Sewickley,
PA.
- Klimasauskas, C. (1993). Making a Difference with Data Transformation.
Advanced Technology for
Developers. June, 1993. High-Tech Communications: Sewickley,
PA.
- Klimasauskas, C. (1993). Simulated Annealing and the Traveling
Salesperson Problem. Advanced
Technology for Developers. June, 1993. High-Tech Communications:
Sewickley, PA.
- Klimasauskas, C. (1993). Some Notes on Genetic Algorithms.
Advanced Technology for
Developers. June, 1993. High-Tech Communications: Sewickley,
PA.
- Klimasauskas, C. (1993). Gray Codes. Advanced
Technology for Developers. May, 1993. High-Tech Communications:
Sewickley, PA.
- Klimasauskas, C. (1993). Genetic Algorithm Optimizes 100-City Route
in 21 Minutes on a PC. Advanced
Technology for Developers. February, 1993. High-Tech Communications:
Sewickley, PA.
- Klimasauskas, C. (1993). Data Analysis for Improved Data-Based
Models. Advanced Technology
for Developers. January, 1993. High-Tech Communications: Sewickley,
PA.
- Klimasauskas, C. (1992). Hybrid Fuzzy Encodings for Improved Back-propagation
Performance. Advanced
Technology for Developers. September, 1992. High-Tech Communications:
Sewickley, PA.
- Klimasauskas, C. (1992). Neural Network's: An Engineering Perspective.
IEEE Communications Magazine. September 1992. pp. 50-53.
- Klimasauskas, C. (1992). Hybrid Technologies: More power for the
Future. Advanced Technology
for Developers. August, 1992. High-Tech Communications: Sewickley,
PA.
- Klimasauskas, C. (1992). Making Fuzzy Logic "Clear".
Advanced Technology for
Developers. May, 1992. High-Tech Communications: Sewickley,
PA.
- Klimasauskas, C. (1991). Applying Neural Networks Part VI: Special
Topics. PC
AI. November / December 1991.
- Klimasauskas, C. (1991). Applying Neural Networks Part V: Integrating
a Trained Network Into an Application. PC AI. September
/ October 1991.
- Guiver, John P.; Klimasauskas, C. (1991). Applying Neural Networks
Part IV: Improving Performance. PC
AI. July / August 1991.
- Klimasauskas, C. (1991). Applying Neural Networks Part III: Training
a Neural Network. PC
AI. May / June 1991.
- Klimasauskas, C. (1991). Applying Neural Networks Part II: A Walk
Through the Application Process. PC
AI. March / April 1991.
- Klimasauskas, C. (1991). Neural Nets Tell Why. Dr.
Dobb's Journal. April 1991.
- Klimasauskas, C. (1991). Applying Neural Networks Part I: An Overview
of the Series. PC
AI. January / February 1991.
- Klimasauskas, C. (1989). Neural Networks: A New Technology for
Information Processing. Heuristics. June 1989.
- Klimasauskas, C. (1988). Neural Networks: A Short Course from Theory
through Application. PC
AI. November / December 1988.
- Klimasauskas, C. (1978). The Economics of Purchasing a Small Computer.
Third West Cost Computer Faire, Palo Alto, CA, November 3-5,
1978.
- Klimasauskas, C.; Ferguson; Henryson. (1974). An Access method
for Direct Access Device Scratch File Management. Technical Report.
Chicago: Argonne National Laboratory.
February 8, 1974.
- Klimasauskas, C. (1973). A Universal Dump Program for Minicomputer
Software Debugging. Technical Report 32-1526, Volume XVI, The
Deep Space Network, Progress Report for May & June 1973. pp.
110-124. Pasadena, California: Jet
Propulsion Laboratory.
- Layland, J. W.; Klimasauskas, C. (1972). A Myopic View of Computer-Based
System Design. Technical Report 32-1526, Volume XIII, The Deep
Space Network, Progress Report for November & December 1972.
pp. 154-167. Pasadena, California: Jet
Propulsion Laboratory.
- Klimasauskas, C. (1972). An Execution Analyzer for the Sigma 5
Computer. Technical Report 32-1526, Volume XII, The Deep Space
Network, Progress Report for September & October 1972. pp. 176-188.
Pasadena, California: Jet Propulsion
Laboratory.
Contact Us Today
- We, at Klimasauskas Group, are highly customer focused. We believe
in delivering value. Our focus is on helping you, the customer, derive
increased value from your existing systems through insightful application
of technology. Sometimes, this involves introducing new solutions. Others,
it may be improving work processes, or using insights from emerging
technologies to improve existing methods. We will work with you in whatever
suits your needs best. Contact us today using the form below.
|
