Author Archives: Linas Vepstas

About Linas Vepstas

Computer Science Researcher - Hanson Robotics

Genetic Crossover in MOSES

MOSES is a system for learning programs from input data.  Given a table of input values, and a column of outputs, MOSES tries to learn a program, the simplest program that can reproduce the output given the input values. The … Continue reading

Posted in Design, Documentation, Introduction, Theory | Tagged , , , , , | 2 Comments

Tuning Metalearning in MOSES

I’ve been studying MOSES recently, with an eye towards performance tuning it. Turns out optimization algorithms don’t always behave the way you think they do, and certainly not the way you want them to. Given a table of values, MOSES … Continue reading

Posted in Design, Development, Documentation, Theory | Tagged , , | 2 Comments

Meaning-Text Theory

During some recent reading, it struck me that a useful framework for thinking about and talking about sentence generation is the MTT or “meaning-text theory” of Igor Mel’cuk, et al Here is one readable reference: Igor A. Mel’čuk and Alain … Continue reading

Posted in Theory | Tagged , , , , , , , , | Leave a comment

Semantic dependency relations

I spent the weekend comparing the Stanford parser to RelEx, and learned a lot. RelEx really does deserve to be called a “semantic relation extractor”, and not just a “dependency relation extractor”. It provides a more abstract, more semantic output … Continue reading

Posted in Design, Development, Documentation, Theory | Tagged , , , , , , , | Leave a comment

Sentence Patterns

I’ve recently resumed work on the question-answering chatbot, and am trying to get it to comprehend a broader range of questions and statements.   The “big idea” is to create a number of “sentence patterns” that the pattern matcher can recognize … Continue reading

Posted in Design, Introduction, Theory | Tagged , , , , , | 2 Comments