<div dir="ltr">Hey Marcus,<br><div><br></div><div>I had updated the proposal for the project <b>Implement Neuroevolution Algorithms</b>. I wanted to know if you had any closing remarks for the same.</div><div><br></div><div>Regards,<br>Nakul</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Mar 17, 2016 at 9:16 PM, Nakul Gulati <span dir="ltr"><<a href="mailto:nakgulati@gmail.com" target="_blank">nakgulati@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hey,<div><br></div><div>Sure I will take a look at the perceptron module, would help me get a better understanding of the code too.</div><div><br></div><div>While read the original paper on NEAT and the implementation of the algorithm with the paper, the author had used pole balancing and XOR problem to benchmark the algorithm. Apart from that I will take a look at the mountain car problem.</div><div><br></div><div>Regards,<br>Nakul</div></div><div class="HOEnZb"><div class="h5"><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Mar 17, 2016 at 8:17 PM, Marcus Edel <span dir="ltr"><<a href="mailto:marcus.edel@fu-berlin.de" target="_blank">marcus.edel@fu-berlin.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div>Hello Nakul,</div><span><div><br></div><div></div><blockquote type="cite"><div>Apologies for getting back so late, I was working on the draft for the proposal.</div><div>I have shared the proposal with mlpack and would appreciate your inputs.</div></blockquote><div><br></div></span><div>Thanks for the proposal, we will take a look on the draft in the next days and</div><div>make comments.</div><span><div><br></div><div></div><blockquote type="cite"><div>At the time Udit implemented the perceptron there wasn't any code to implement</div><div>the perceptron using the FNN class.</div><div><br></div><div>If there is a need to rewrite the perceptron using the fnn interface, I would</div><div>work on that too.</div></blockquote><div><br></div></span><div>If you like you can rewrite the perceptron using the ann modules.</div><span><div><br></div><div></div><blockquote type="cite"><div>After your mention of the pole balancing problem, I read more about it and it is</div><div>regarded as a pseudo-standard benchmark test for neuroevolution algorithms, so I</div><div>would implement that as part of testing of the algorithm.</div></blockquote><div><br></div></span><div>Sounds great, there are a couple of other tests that could be interessting like:</div><div>the mountain car task.</div><div><br></div><div>Thanks,</div><div>Marcus</div><div><div><div><br></div><div><br></div><div><blockquote type="cite"><div>On 17 Mar 2016, at 14:11, Nakul Gulati <<a href="mailto:nakgulati@gmail.com" target="_blank">nakgulati@gmail.com</a>> wrote:</div><br><div><div dir="ltr">Hey Marcus,<div><br></div><div>Apologies for getting back so late, I was working on the draft for the proposal. I have shared the proposal with mlpack and would appreciate your inputs.</div><div><br></div><div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">At the time Udit implemented the perceptron there wasn't any code to implement<br>the perceptron using the FNN class.</blockquote><div><br></div><div>If there is a need to rewrite the perceptron using the fnn interface, I would work on that too. </div></div><div><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">Or the other way. Take a look at the pole balancing problem that could be a neat<br>test case.</blockquote><div><br></div><div>After your mention of the pole balancing problem, I read more about it and it is regarded as a pseudo-standard benchmark test for neuroevolution algorithms, so I would implement that as part of testing of the algorithm.</div><div><br></div><div>Regards,<br>Nakul</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Sun, Mar 13, 2016 at 8:25 PM, Marcus Edel <span dir="ltr"><<a href="mailto:marcus.edel@fu-berlin.de" target="_blank">marcus.edel@fu-berlin.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div>Hello Nakul,</div><span><div><br></div><div></div><blockquote type="cite"><div>I have gone through the literature and have a better understanding of the</div><div>algorithms and also have been spending time with the source code. It took a</div><div>little time but now I am acquainted with the source code. After understanding</div><div>the ann code I went ahead to check out the perceptron code to see it in action</div><div>but to my surprise perceptron although being a feed forward single layer neural</div><div>network doesn't use fnn code, is there a particular reason for it, am I missing</div><div>something?</div></blockquote><div><br></div></span><div>At the time Udit implemented the perceptron there wasn't any code to implement</div><div>the perceptron using the FNN class.</div><span><div><br></div><div></div><blockquote type="cite"><div>Also about the GSoC project I feel confident with CNE and NEAT but like you had</div><div>mentioned the importance of test cases I would the priority would be to</div><div>implement CNE with test cases and then move to NEAT.</div></blockquote><div><br></div></span><div>Or the other way. Take a look at the pole balancing problem that could be a neat</div><div>test case.</div><div><br></div><div>I hope this is helpful, let me know if I can clarify anything.</div><div><br></div><div>Thanks,</div><div>Marcus</div><div><div><div><br></div><div><br></div><div><blockquote type="cite"><div>On 13 Mar 2016, at 11:53, Nakul Gulati <<a href="mailto:nakgulati@gmail.com" target="_blank">nakgulati@gmail.com</a>> wrote:</div><br><div><div dir="ltr">Hey Marcus,<div><br></div><div>I have gone through the literature and have a better understanding of the algorithms and also have been spending time with the source code. It took a little time but now I am acquainted with the source code. After understanding the ann code I went ahead to check out the perceptron code to see it in action but to my surprise perceptron although being a feed forward single layer neural network doesn't use fnn code, is there a particular reason for it, am I missing something?</div><div><br></div><div>Also about the GSoC project I feel confident with CNE and NEAT but like you had mentioned the importance of test cases I would the priority would be to implement CNE with test cases and then move to NEAT.</div><div><br></div><div>Regards,<br>Nakul Gulati</div><div><a href="https://nakulgulati.com/" target="_blank">Website</a> || <a href="http://in.linkedin.com/in/nakulgulati" target="_blank">LinkedIn</a> || <a href="https://github.com/nakulgulati/" target="_blank">GitHub</a></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Mar 10, 2016 at 7:17 PM, Marcus Edel <span dir="ltr"><<a href="mailto:marcus.edel@fu-berlin.de" target="_blank">marcus.edel@fu-berlin.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div>Hello Nakul,</div><div><br></div><div>It is a good decision to start with something easy that could be CNE or NEAT.</div><div>Afterwards, we can use that as a baseline for comparison with other</div><div>implementation. You should keep in mind, that you have to write tests, for every</div><div>evolution algorithm you write, and that often takes more time than the actual</div><div>implementation. Anyway, this year google made draft proposals part of the</div><div>proposal workflow. So we can comment on your timeline etc. and give feedback</div><div>once you submitted your application. Also take a look at:</div><div><a href="http://write.flossmanuals.net/gsocstudentguide/" target="_blank">http://write.flossmanuals.net/gsocstudentguide/</a></div><div><br></div><div>I hope this is helpful. Let me know if I can clarify anything,</div><div><br></div><div>Thanks,</div><div>Marcus</div><div><br></div><div><blockquote type="cite"><div><div><div>On 10 Mar 2016, at 04:12, Nakul Gulati <<a href="mailto:nakgulati@gmail.com" target="_blank">nakgulati@gmail.com</a>> wrote:</div><br></div></div><div><div><div><div dir="ltr"><div>Hi,</div><div><br></div><div>I am a final year Computer Science and Engineering student. I am interested in contributing to the project <b>Neuroevolution algorithms</b> as part of GSoC 2016. Some of the courses completed which are relevant to this project are: Data Structures and Design and Analysis of Algorithms, Soft Computing and Artificial Neural Networks.</div><div><br></div><div>In order to get better understanding about the project and mlpack, the following steps were taken:</div><div><ul><li>Compiled and tested mlpack on OS X</li><li>Compiled and tested nes emulator communication code: during this I ended up crashing the emulator hosted on <a href="http://mario.urgs.org/" target="_blank">mario.urgs.org</a> 4561, <a href="https://github.com/zoq/nes/issues/1" target="_blank">link to issue</a>. Which was promptly fixed by Marcus Edel, big shout out to him.</li></ul><div>Currently I am studying the reading material, starting with the CNE algorithm.</div></div><div><br></div><div>For scope of GSoC project I propose to start with the implementation of the CNE algorithm at first and then move to a second (name would be specified soon) if time permits.</div><div><br></div><div>I have little trouble setting up an exact timeline for the project. I would love to hear your insight about the approach and timeline and if you think what I propose is realistic.</div><div><br></div>-- <br><div><div dir="ltr"><div><div dir="ltr"><div dir="ltr">Regards,<br>Nakul Gulati</div><div><a href="https://nakulgulati.com/" target="_blank">Website</a> || <a href="http://in.linkedin.com/in/nakulgulati" target="_blank">LinkedIn</a> || <a href="https://github.com/nakulgulati/" target="_blank">GitHub</a></div></div></div></div></div>
</div></div></div>
_______________________________________________<br>mlpack mailing list<br><a href="mailto:mlpack@cc.gatech.edu" target="_blank">mlpack@cc.gatech.edu</a><br><a href="https://mailman.cc.gatech.edu/mailman/listinfo/mlpack" target="_blank">https://mailman.cc.gatech.edu/mailman/listinfo/mlpack</a><br></div></blockquote></div><br></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div><div dir="ltr"><div><div dir="ltr"><div dir="ltr">Regards,<br>Nakul Gulati</div><div><a href="https://nakulgulati.com/" target="_blank">Website</a> || <a href="http://in.linkedin.com/in/nakulgulati" target="_blank">LinkedIn</a> || <a href="https://github.com/nakulgulati/" target="_blank">GitHub</a></div></div></div></div></div>
</div>
</div></blockquote></div><br></div></div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div><div dir="ltr"><div><div dir="ltr"><div dir="ltr">Regards,<br>Nakul Gulati</div><div><a href="https://nakulgulati.com/" target="_blank">Website</a> || <a href="http://in.linkedin.com/in/nakulgulati" target="_blank">LinkedIn</a> || <a href="https://github.com/nakulgulati/" target="_blank">GitHub</a></div></div></div></div></div>
</div>
</div></blockquote></div><br></div></div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div><div dir="ltr"><div><div dir="ltr"><div dir="ltr">Regards,<br>Nakul Gulati</div><div><a href="https://nakulgulati.com" target="_blank">Website</a> || <a href="http://in.linkedin.com/in/nakulgulati" target="_blank">LinkedIn</a> || <a href="https://github.com/nakulgulati/" target="_blank">GitHub</a></div></div></div></div></div>
</div>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div dir="ltr">Regards,<br>Nakul Gulati</div><div><a href="https://nakulgulati.com" target="_blank">Website</a> || <a href="http://in.linkedin.com/in/nakulgulati" target="_blank">LinkedIn</a> || <a href="https://github.com/nakulgulati/" target="_blank">GitHub</a></div></div></div></div></div>
</div>