Nondeterministic States Research Articles

SiamCMN: Jointing cycle memory and correlation network for Siamese object tracking

In heterogeneous scenes with latent non-deterministic states, template features determine the performance of the Siamese network-based trackers, however, background noise is easily introduced in the search area matching process, which causes the trackers lack sufficient discrimination. To solve this problem, this paper proposes an object tracking algorithm based on the cyclic memory module. Firstly, the extracted primary features are mapped to the cyclic spatial memory network, the spatial memory unit is used to store the interference components. On this basis, the cyclic structure is used to filter out the interference components and refine the template features. Then a channel correlation network is introduced into the proposed method, which can mine semantic complementary information and capture global contextual relevance by modeling the positive and the negative correlation between the channels. The experimental results on tracking benchmarks such as OTB50, OTB100, UAV123, VOT2018, GOT10k, and LaSOT have shown that the proposed method outperforms recent representative algorithms in tracking accuracy in challenging scenarios.

Using a process algebra interface for verification and validation of UML statecharts

UML diagrams are the conventional methods for visual modeling systems. Among them, the Statechart diagrams are used to show the runtime behavior of a system, but the correctness of such diagrams is the primary concern of the designers because of concurrency issues like livelock, inaccessible states, and non-deterministic states. Process algebra methods have the capabilities that are suitable for verification and validation of Statecharts. To this end, in this paper, process algebra language LOTOS (Language Of Temporal Ordering Specification) is used as the target language, and a method is presented to map UML Statecharts to the LOTOS processes, called USLP. Then the correctness of the proposed mappings is proved by demonstrating the isomorphism relation between the Labeled Transition System (LTS) of a Statechart and the LTS of its transformed LOTOS specification. Next, tools CADP (Construction and Analysis of Distributed Processes) is used for verification and validation of the mapped LOTOS models, and the CSP process algebra and its tools, FDR are used to verify the properties could not be verified by the LOTOS and its toolset. The experimental results show our approach can: (1) verify some properties (the issues) that are not verified by other approaches and (2) reduce the space that should be searched to verify the properties.

Uncertain production scheduling optimization in open-pit mines and its ellipsoidal robust counterpart

The block extraction sequence is among the most challenging and important issues that should be regarded through whole mining operations to reach the maximum amount of profit. Such open-pit production scheduling problems can be solved under either deterministic or non-deterministic states. This study aims to model the problem applying robust counterpart linear optimization that uses the ellipsoidal set based counterpart. Block economic value as the objective function coefficient, and block weights and operational capacity as the constraints coefficients, are considered as uncertainty sources. Besides the exact mathematical modelling, a genetic algorithm is applied to solve the nonlinear ellipsoidal counterpart. The model, which was solved in deterministic and uncertain conditions, terminated in results with some differences in scheduled plans. The price of robustness is obtained at different conservative levels.

Mathematical Foundations of Probabilistic Estimation in Fuzzy Bayesian Belief Networks

Specific problems arising in the creation of Bayesian networks for processing fuzzy information are considered. Conditions are obtained that ensure the probabilistic correctness of a priori data and results of network computing. For the case of nondeterministic states of network vertices, a fuzzy linear interpolation procedure is proposed. The presented results make it possible to perform correct probabilistic Bayesian estimation over fuzzy networks of any configuration.

Probability propagation in fuzzy Bayesian belief networks with nondeterministic states

The conceptual basis of fuzzy Bayesian belief networks with nondeterministic states is considered. The concept of a fuzzy probability estimate as a fuzzy relation of special type is introduced and its geometrical interpretation is given. Functional transformations of fuzzy probability estimates are defined and a multidimensional linear interpolation procedure is developed. Fundamental aspects of information distribution in fuzzy Bayesian belief networks with nondeterministic states are considered.

On some variations of two-way probabilistic finite automata models

Rabin [M. Rabin, Probabilistic finite automata, Information and Control (1963) 230–245] initiated the study of probabilistic finite automata (pfa). Rabin’s work showed a crucial role of the gap in the error bound (for accepting and non-accepting computations) in the power of the model. Further work resulted in the identification of qualitatively different error models (one-sided error, bounded and unbounded errors, no error etc.) Karpinski and Verbeek [M. Karpinski, R. Verbeek, There is no polynomial deterministic simulation of probabilistic space with two-way random-tape generator, Information and Control 67 (1985) 158–162] and Nisan [N. Nisan, On read-once vs. multiple access to randomness in logspace, in: Proc. of Fifth IEEE Structure in Complexity Theory, Barcelona, Spain, 1990, pp. 179–184] studied a model of probabilistic automaton in which the tape containing random bits can be read by a two-way head. They presented results comparing models with one-way vs. two-way access to randomness. Dwork and Stockmeyer [C. Dwork, L. Stockmeyer, Interactive proof systems with finite state verifiers, IBM Report RJ 6262, 1988] and Condon et al. [A. Condon, et al., On the power of finite automata with both nondeterministic and probabilistic states, SIAM Journal on Computing (1998) 739–762] studied a model of 2-pfa with nondeterministic states (2-npfa). In this paper, we present some results about the above mentioned variations of probabilistic finite automata, as well as a model of 2-pfa augmented with a pebble studied in [B. Ravikumar, Some observations on two-way probabilistic finite automata, in: Proc. of the Foundations of Software Technology and Theoretical Computer Science, 1992, pp. 392–403]. Our observations indicate that these models exhibit subtle variations in their computational power. We also mention many open problems about these models. Complete characterizations of their power will likely provide deeper insights about the role of randomness in space-bounded computations.

On the Structure of Logspace Probabilistic Complexity Classes

We investigate hierarchical properties and logspace reductions of languages recognized by logspace probabilistic Turing machines, Arthur--Merlin games, and games against nature with logspace probabilistic verifiers. Each logspace complexity class is decomposed into a hierarchy based on corresponding two-way multihead finite-state automata and we (eventually) prove the separation of the hierarchy levels (even for languages over a single-letter alphabet); furthermore, we show efficient reductions of each logspace complexity class to, or between, low levels of its corresponding hierarchy. We find probabilistic and probabilistic-plus-nondeterministic variants of Savitch's maze threading problem which are logspace complete for PL (the class of languages recognized by logspace probabilistic Turing machines) and, respectively, PPP (the class of languages recognized by polynomial-time deterministic Turing machines), and which can be recognized by one-way non-sensing two-head (or one-way one-head one-counter) finite-state automata with probabilistic and both probabilistic and nondeterministic states, respectively.