Solving Equation Systems by Agreeing and Learning
- Cite this paper as:
- Schilling T.E., Raddum H. (2010) Solving Equation Systems by Agreeing and Learning. In: Hasan M.A., Helleseth T. (eds) Arithmetic of Finite Fields. WAIFI 2010. Lecture Notes in Computer Science, vol 6087. Springer, Berlin, Heidelberg
We study sparse non-linear equation systems defined over a finite field. Representing the equations as symbols and using the Agreeing algorithm we show how to learn and store new knowledge about the system when a guess-and-verify technique is used for solving. Experiments are then presented, showing that our solving algorithm compares favorably to MiniSAT in many instances.
Keywordsagreeing multivariate equation system SAT-solving dynamic learning
Unable to display preview. Download preview PDF.