Reinforced walk on graphs and neural networks

• Autorzy: Józef Myjak , Ryszard Rudnicki
• Języki publikacji: EN

Abstrakty

EN

A directed-edge-reinforced random walk on graphs is considered. Criteria for the walk to end up in a limit cycle are given. Asymptotic stability of some neural networks is shown.

Kategorie tematyczne

• 60K37: Processes in random environments
• 60J20: Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.)
• 60G50: Sums of independent random variables; random walks
• 60J10: Markov chains (discrete-time Markov processes on discrete state spaces)
• 82C41: Dynamics of random walks, random surfaces, lattice animals, etc.

• Wydawca: Institute of Mathematics Polish Academy of Sciences
• Czasopismo: Studia Mathematica
• Rocznik: 2008
• Tom: 189
• Numer: 3
• Strony: 255-268

Daty

wydano

2008

Twórcy

autor

Józef Myjak

• Dipartimento di Matematica Pura ed Applicata, Università di L'Aquila, Via Vetoio, 67100 L'Aquila, Italy
• WMS AGH, al. Mickiewicza 30, 30-059 Kraków, Poland

autor

Ryszard Rudnicki

• Institute of Mathematics, Polish Academy of Sciences, Bankowa 14, 40-007 Katowice, Poland
• Institute of Mathematics, Silesian University, Bankowa 14, 40-007 Katowice, Poland

Identyfikatory

DOI

10.4064/sm189-3-4