arXiv subjects
Publications and source records attributed to Cewei Cui.
We study random words in a weighted regular language that achieve the maximal free energy using thermodynamics formalism. In particular, typical words in the language are algorithmically generated which have applications in computer security (anomaly detection) and software enegineering (test case generation).
This paper develops two heuristic algorithms to solve graph isomorphism, using free energy encoding. The first algorithm uses four types of encoding refinement techniques such that every graph can be distinguished by a canonical number computed by the algorithm. The second algorithm injects energy into the graph to conduct individualization such that the correspondence relation between a pair of isomorphic graphs can be found. The core principle behind the two algorithms is encoding discrete structures as real numbers. A large set of experiments demonstrated the effectiveness of our algorithms.
This paper develops a free energy theory from physics including the variational principles for automata and languages and also provides algorithms to compute the energy as well as efficient algorithms for estimating the nondeterminism in a nondeterministic finite automaton. This theory is then used as a foundation to define a semantic similarity metric for automata and languages. Since automata are a fundamental model for all modern programs while languages are a fundamental model for the programs' behaviors, we believe that the theory and the metric developed in this paper can be further used for real-word programs as well.
A program can be considered as a device that generates discrete time signals, where a signal is an execution. Shannon information rate, or bit rate, of the signals may not be uniformly distributed. When the program is specified by a finite state transition system, algorithms are provided in identifying information-rich components. For a black-box program that has a partial specification or does not even have a specification, a bit rate signal and its spectrum are studied, which make use of data compression and the Fourier transform. The signal provides a bit-rate coverage for testing the black-box while its spectrum indicates a visual representation for execution's information characteristics.