Zobrazeno 1 - 10
of 17
pro vyhledávání: '"Phansalkar VV"'
Autor:
Phansalkar, VV, Thathachar, MAL
Publikováno v:
IndraStra Global.
This paper analyses the behaviour of a general class of learning automata algorithms for feedforward connectionist systems in an associative reinforcement learning environment. The type of connectionist system considered is also fairly general. The a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=issn23813652::a58edc3d9ee8de7e277c5c040f9941ec
https://igi.indrastra.com/items/show/196131
https://igi.indrastra.com/items/show/196131
Autor:
Thathachar, MAL, Phansalkar, VV
Learning algorithms for feedforward connectionist systems in a reinforcement learning environment are developed and analyzed in this paper. The connectionist system is made of units of groups of learning automata. The learning algorithm used is the $
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::a64c788454d81243ce4be8644367970b
http://eprints.iisc.ernet.in/9812/
http://eprints.iisc.ernet.in/9812/
Autor:
Phansalkar, VV, Thathachar, MAL
Publikováno v:
IndraStra Global.
Analyzes the long-term behavior of the REINFORCE and related algorithms (Williams, 1986, 1988, 1992) for generalized learning automata (Narendra and Thathachar, 1989) for the associative reinforcement learning problem (Barto and Anandan, 1985). The l
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=issn23813652::e6b2a924d040bc46c61759e8a3b47a34
https://igi.indrastra.com/items/show/194659
https://igi.indrastra.com/items/show/194659
Autor:
Thathachar, MAL, Phansalkar, VV
Publikováno v:
IndraStra Global.
A feedforward network composed of units of teams of parameterized learning automata is considered as a model of a reinforcement teaming system. The internal state vector of each learning automaton is updated using an algorithm consisting of a gradien
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=issn23813652::b1ca32b2bd8131ff14195148499910c3
https://igi.indrastra.com/items/show/196147
https://igi.indrastra.com/items/show/196147
Autor:
Borkar , VS, Phansalkar, VV
For a class of distributed recursive algorithms, it is shown that a stochastic approximation-like tapering stepsize routine suppresses the effects of interprocessor delays.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::195c15a93bf8f09413230873eb1d4dc7
http://eprints.iisc.ernet.in/36630/
http://eprints.iisc.ernet.in/36630/
Autor:
Phansalkar, VV, Thathachar, MAL
Publikováno v:
IndraStra Global.
A feedforward network composed of units of teams of parametrised learning autmata is considered as a mode2 of a reinforcement learning system. The parameters of each learning automaton are updated using an algorithm consisting of a gradient following
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::1424d97e6d2d0fa802ccd9fda3eb8a47
http://eprints.iisc.ac.in/7117/1/36.pdf
http://eprints.iisc.ac.in/7117/1/36.pdf
Autor:
Thathachar, MAL, Phansalkar, VV
Publikováno v:
IndraStra Global.
A model made of units of teams of learning automata is developed for the three layer pattern classifier. The algorithm is approximated by an ODE, using weak convergence methods. The pattern recognition problem is posed as a constrained maximization p
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::a8cebdfb97ccc4ed9fe6115fdc724174
https://igi.indrastra.com/items/show/194636
https://igi.indrastra.com/items/show/194636
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.