Mail of Readers of the pagina Following Pagina Previous Pagina In view of the great amount of correspondence received by the pagina some of mails are published aqui as a example received. In order to respect the privacy of the senders, the last names and e-mails have erased and the christian names of such are left only. Nevertheless, I do not like the anonymous post office asi that if you send one please she adds his last name to the same one. Some representative answers are included in the pagina, others have been made individually, but in general I have tried to respond to all the messages. I have noticed that several mails can be grouped in categorias, since they make reference to same the thematic one. For them, it seems better to give them a global answer me. Mails congratulating or saluting Lea I deeply thank for to the readers the greetings. Mails communicating that will become use of my work: They are mails par excellence for which the mail of readers was created. If you are making one monografia, thesis or similars, can copy definitions or paragraphs of this work for his, but it must identify in its work to the author (that is to my) and the in question fact of data extragone of this work. The considered people but send mails to notify to me that they used my work as mentions bibliografica of his. I thank for east type of mails. Mails asking for bibliografia on the subject: Bibliografia used and of consultation is published in the work. In general I publish what I have available. Leave from the base that I do not have but that what there he appears. Mails asking for software recommendations: I am not a specialist in test of software. This means that I have not proven all the softwares available on networks nor on AG as to advise on as they are best nor but the cheap ones. In the sites of download of the network there are some programs (payments, and gratuitous) to lower. Nevertheless, I abstain to think in an u others please. If the readers wish to send mails praising or defenestrando algun software I do not have problems in publishing it. Mails doing precise questions: It depends on the type of question treatment to answer them all. Mails requesting the content of the done professional works: The done professional works to order are property of the people who have ordered them. They straight keep copyright or from author on such, and therefore they cannot be published. Mails general requesting aid for the works that the readers are doing: If you the work is taken to read the listing of mails, will observe that much people request aid for the works of investigation that are making. When I began to publish my works in the network, the idea was the one to make a shared in common contribution to the community of university students and postdegree of low resources that have difficulty to accede to the bibiography on Artificial intelligence. Another objective was the one to contribute with bibliography in Castilian in subjects that are handled generally in English. Personally I am a person who must work to gain the life. Therefore, all my labor useful time it use in the accomplishment of my professional tasks, which prevent to deeply dedicate me me but to the site. The page, on the contrary, is a public and gratuitous service for the community of Internet, and it does not have patrocinantes subventions nor of any species, reason why it does not generate income nor I can live on her. Banners is put by the proprietors of the Hosting (gratuitous storage). Therefore, in general I must clarify to the readers who the gratuitous consultancy on IA does not comprise of the offered services. A last reflection that goes up to around the head to me with respect to this subject is the following one: Throughout my professional life I have asked for aid and shared works with other colleagues, but in all the cases I have recognized them the Co-responsibility of such when the aid is important, or the mention of collaborator in the prologue of the work when the contribution is smaller. The mentioned authors are exposed in the bibliography. All aid which I have requested to the aims of accomplishment or publication of material has been notified and recognized clearly. Nevertheless, I have seen that the totality of orders that arrive to me speaks of the necessity of aid or advising that has each person, but no express but minimum intencíon to make the actions before happiness, reason why single it reduces to think that really what they wish is the aid of my part, without intentions of recognition some to me about the made investigations... As I already said, when I need aid I do not handle myself of that way. Listing of last received Mail: Hello, my name is Oscar B, is of Chile and is attending the last year of Engineering in Computer science in the Professional Institute Inacap. This semester I must give my memory and the subject that I will treat is the neuronal networks. Visisté your page and ví your curriculum, I see that or you have been a certain time in the subject so it would thank for much to you if me you could facilitate information about neuronal networks, or are directions of some pages, bibliography, some experience yours in the study of the subject, etc. any thing will be welcome. I hope to have a good welcome. Ciao LIKE THESE MARCELO, I WANT DECIRTE THAT LAW TU PAGINA IN INTERNET OF NEURONAL NETWORKS, me VERY GOOD PARECIO and I HELP MUCHISIMO TO INCLUDE/UNDERSTAND THEM. I WANT INFORMARTE THAT I AM WORKING A THESIS ABOUT THIS TEMA.LO THAT I TRY TO DO IS A SIMULATION OF A NEURONAL NETWORK WORKING TO SOLVE ANY PROBLEM. IN SPECIAL ELECTRONICO. I MAKE SEMESTRAL DECIMO OF INGENIERIA ELECTRONICA IN UNIVERSITY AUTONOMA OF THE CARIBBEAN OF OCLcOcMcBcIcA(barranquilla). AS PODRAS TO UNDERSTAND I AM DOING MY THESIS OF DEGREE and I WANT THAT YOU COLLABORATE WITH SUGGESTIONS Or a AYUDA.AQUI to ME I SEND to MY mail xxxxxx@YAHOO.ES ESCRIBEME SOON. SO LONG AND THANKS. My name is Jhon Alexander, is Colombian of 22 years, student of tenth semester of Ingenieria de Sistemas, is looking for in that making my project of degree and because one of the areas that my university works is the neuronal networks. So far single I am documentandome and the reason of looking for on the subject in consideration is that there has been no a project dedicated to this technology, parecio quite important to look for to me in Internet and to know something on the matter. Beforehand I want to congratulate it by its excellent work published in the Web, very easy to understand, so far single I will use it like general information, soon if I decide to make my project in this area I believe that I will be bothering it when some restlessness appear. Thanks again. Dear Marcelo, analyzing its coarse experience in desarollo of software and especificamente its works of investigation with relacionacion to the neuronal networks, him agradeceria if it gives his recommendations to me on an economic software and of facil implementation for the prediction of a univariada series of time. Beforehand I appreciate its collaboration and direction to him on the matter. MR.: MARCELO GONZALEZ B. Hello my name is Patricia, I am Statistical Engineer, at the moment course 2° year of the Magíster in Management Mention Accounting and Finances in the Catholic University of Valparaiso of Chile. My interest in writing to him is because I try to make my thesis on the subject of neuronal networks in the prediction of actions, and in addition in my work we are going it to use in the analysis of frauds in the prices. It wanted to know that it thinks on the subject that I am going to treat since just it study studying, and in addition I would like to know in that bibliography could support to me One takes leave kindly of You. Patricia. Lic. Marcelo González Considered Forest My name is Reynaldo, I am establishing communication with You from the city of Arequipa Peru, good when reading yours I articulate on RNA and other but that appears in the network, I have some doubts, which to him immensely You would thank for you can aclararmelas. - In the case of having continuous and nonbinary data (exits and entrances), that type of network and which activation function would use, I have been modeling a network and the truth that when using the sigmoide function I see myself with the surprise that the training stops because the values reach 0 or 1, - I can apply backpropagation by lots and each patron, is to say if I have rule delta, I do this same for each patron of entrance-exit - the weights of the thresholds also are updated? - really single I have of the information spilled in Internet and algun book which does not describe ningun example on the generalization of the rule delta. You could illustrate to me in detail but on this type of training - With respect to the previous thing, the error is the difference between the wished exit and the obtained exit, as I calculate the difference wished for each hidden neuron. Muchisimas Gracias Sincerely Reynaldo NATIONAL UNIVERSITY SAN AGUSTIN 2da SPECIALTY IN INGENIERIA OF PRODUCTION Marcelo Gonzalez: I am student of engineering in computation and computer science and have a work of the subjet of ' theory of autómatas' with respect to the following complex systems: - Cellular Robots. - Neuronal Networks. - Blurred Robots. - Genetic Algorithms. - programmable Robots in the industrial process control. He consults: You have material on him subject, or some direction that could suggest to me. From already I appreciate its attention. Atte. Andrea Santiago Chile.