Learning Finite-State Transducers: Evolution Versus Heuristic State Merging

Lucas, Simon M.; Reynolds, T.J.

doi:10.1109/tevc.2006.880329

Cited by 14 publications

(15 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this research, we followed the work of [4] and used two tables to represent a FST. The first table is transition table (see Figure V), which gives the next state of a FST based on the input symbol and the current state.…”

Section: Evolving Finite State Transducersmentioning

confidence: 99%

“…This system only uses mutation operator, which works the same way as that in [4]. First, a decision is made with equal probability to either mutate the transition table or the output table.…”

Section: B the Evolutionary Systemmentioning

confidence: 99%

“…When an entry is modified, a symbol is chosen from a uniform distribution of all possible symbols except the current symbol. According to [4], a single call to the mutation operator is most likely to produce one or two changes to the FST tables, but can also produce more.…”

Section: B the Evolutionary Systemmentioning

confidence: 99%

See 2 more Smart Citations

Evolving Finite State Transducers to Interpret Deepwater Reservoir Depositional Environments

Wilkinson

Clark

et al. 2008

2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence)

View full text Add to dashboard Cite

Predicting oil recovery efficiency of deepwater reservoirs is a challenging task. One approach to characterize and predict the producibility of a reservoir is by analyzing its depositional information. In a deposition-based stratigraphic interpretation framework, one critical step is the identification and labeling of the stratigraphic components in the reservoir according to their depositional information. This interpretation process is labor intensive and can produce different results depending on the stratigrapher who performs the analysis. To relieve stratigrapher's workload and to produce more consistent results, this research developed a methodology to automate this process using various computational intelligent techniques. Using a well log data set, we demonstrated that the developed methodology and the designed workflow can produce finite state transducer models that interpret deepwater reservoir depositional environments adequately.

show abstract

Section: Evolving Finite State Transducersmentioning

confidence: 99%

Section: B the Evolutionary Systemmentioning

confidence: 99%

See 1 more Smart Citation

Evolving Finite State Transducers to Interpret Deepwater Reservoir Depositional Environments

Wilkinson

Clark

et al. 2008

2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence)

View full text Add to dashboard Cite

show abstract

“…Ngom et al [7] used genetic simulation for Moore machine identification, Tongchim and Chongstitvatana [8] investigated parallel implementation of the GA to solve the problem of FSM synthesis. Lucas [9] paid more attention to finite state transducers and he and Reynolds [10] compared this method to 'Heuristic State Merging'. Niparnan and Chongstitvatana [11] improved GA by evolving only the state transition function.…”

Section: Introductionmentioning

confidence: 99%

An approach to the inference of finite state machines based on a gravitationally-inspired search algorithm

Spichakova¹

2013

Proc. Estonian Acad. Sci.

View full text Add to dashboard Cite

As the inference of a finite state machine from samples of its behaviour is NP-hard, heuristic search algorithms need to be applied. In this article we propose a methodology based on applying a new gravitationally-inspired heuristic search algorithm for the inference of Moore machines. Binary representation of a Moore machine, an evaluation function, and the required parameters of the algorithm are presented. The experimental results show that this method has a lot of potential.

show abstract

“…There are two major ways to define a fitness function: based on comparing the FSM's behavior to a standard recorded in test examples [3], [6], [7] or based on modeling in some environment [8], [9]. Both of these ways may lead to high computational costs of a single fitness function evaluation, which directly leads to an increased cost of building an optimal FSM.…”

Section: Introductionmentioning

confidence: 99%

Learning Finite-State Machines: Conserving Fitness Function Evaluations by Marking Used Transitions

Chivilikhin

Ulyantsev

2013

2013 12th International Conference on Machine Learning and Applications

View full text Add to dashboard Cite

This paper is dedicated to the problem of learning finite-state machines (FSMs), which plays a key role in automatabased programming. Metaheuristic algorithms commonly applied to this problem often use FSM mutations (small changes in the FSM structure) for solution construction. Most of them do not employ the specifics of FSMs in their work. We propose a new simple method for improving performance of these algorithms. The basic idea is to mark those transitions of FSMs that were used during fitness evaluation. Then, if a FSM mutation changes a transition that was not used in fitness evaluation, the fitness function value need not be calculated for the mutated FSM. This observation allows to conserve fitness evaluations, which often have high computational costs. The proposed method has been incorporated into several traditional and recent FSM learning algorithms based on evolutionary strategies, genetic algorithms and ant colony optimization. Experimental results are reported showing that the new method significantly improves performance of two methods based on evolutionary strategies and ant colony optimization.

show abstract

Learning Finite-State Transducers: Evolution Versus Heuristic State Merging

Cited by 14 publications

References 35 publications

Evolving Finite State Transducers to Interpret Deepwater Reservoir Depositional Environments

Evolving Finite State Transducers to Interpret Deepwater Reservoir Depositional Environments

An approach to the inference of finite state machines based on a gravitationally-inspired search algorithm

Learning Finite-State Machines: Conserving Fitness Function Evaluations by Marking Used Transitions

Contact Info

Product

Resources

About