Thanks Tim. It's frustrating to see the community has very little experience with MLN, after all, this is the smartest group of people I know in computer science. Okay, the focus here will be on code.
On Friday, September 2, 2016 at 1:16:56 PM UTC-3, Viral Shah wrote: > > I agree with John here. This is totally unacceptable, and is making the > experience poorer for others. > > -viral > > On Friday, September 2, 2016 at 8:48:44 PM UTC+5:30, John Myles White > wrote: >> >> May I also point out to the My settings button on your top right corner > >>> My topic email subscriptions > Unsubscribe from this thread, which would've >>> spared you the message. >> >> >> I'm sorry, but this kind of attitude is totally unacceptable, Kevin. I've >> tolerated your misuse of the mailing list, but it is not acceptable for you >> to imply that others are behaving inappropriately when they complain about >> your unequivocal misuse of the mailing list. >> >> --John >> >> On Friday, September 2, 2016 at 7:23:27 AM UTC-7, Kevin Liu wrote: >>> >>> May I also point out to the My settings button on your top right corner >>> > My topic email subscriptions > Unsubscribe from this thread, which >>> would've spared you the message. >>> >>> On Friday, September 2, 2016 at 11:19:42 AM UTC-3, Kevin Liu wrote: >>>> >>>> Hello Chris. Have you been applying relational learning to your Neural >>>> Crest Migration Patterns in Craniofacial Development research project? It >>>> could enhance your insights. >>>> >>>> On Friday, September 2, 2016 at 6:18:15 AM UTC-3, Chris Rackauckas >>>> wrote: >>>>> >>>>> This entire thread is a trip... a trip which is not really relevant to >>>>> julia-users. You may want to share these musings in the form of a blog >>>>> instead of posting them here. >>>>> >>>>> On Friday, September 2, 2016 at 1:41:03 AM UTC-7, Kevin Liu wrote: >>>>>> >>>>>> Princeton's post: >>>>>> http://www.nytimes.com/2016/08/28/world/europe/france-burkini-bikini-ban.html?_r=1 >>>>>> >>>>>> Only logic saves us from paradox. - Minsky >>>>>> >>>>>> On Thursday, August 25, 2016 at 10:18:27 PM UTC-3, Kevin Liu wrote: >>>>>>> >>>>>>> Tim Holy, I am watching your keynote speech at JuliaCon 2016 where >>>>>>> you mention the best optimization is not doing the computation at all. >>>>>>> >>>>>>> Domingos talks about that in his book, where an efficient kind of >>>>>>> learning is by analogy, with no model at all, and how numerous >>>>>>> scientific >>>>>>> discoveries have been made that way, e.g. Bohr's analogy of the solar >>>>>>> system to the atom. Analogizers learn by hypothesizing that entities >>>>>>> with >>>>>>> similar known properties have similar unknown ones. >>>>>>> >>>>>>> MLN can reproduce structure mapping, which is the more powerful type >>>>>>> of analogy, that can make inferences from one domain (solar system) to >>>>>>> another (atom). This can be done by learning formulas that don't refer >>>>>>> to >>>>>>> any of the specific relations in the source domain (general formulas). >>>>>>> >>>>>>> Seth and Tim have been helping me a lot with putting the pieces >>>>>>> together for MLN in the repo I created >>>>>>> <https://github.com/hpoit/Kenya.jl/issues/2>, and more help is >>>>>>> always welcome. I would like to write MLN in idiomatic Julia. My >>>>>>> question >>>>>>> at the moment to you and the community is how to keep mappings of >>>>>>> first-order harmonic functions type-stable in Julia? I am just >>>>>>> getting acquainted with the type field. >>>>>>> >>>>>>> On Tuesday, August 9, 2016 at 9:02:25 AM UTC-3, Kevin Liu wrote: >>>>>>>> >>>>>>>> Helping me separate the process in parts and priorities would be a >>>>>>>> lot of help. >>>>>>>> >>>>>>>> On Tuesday, August 9, 2016 at 8:41:03 AM UTC-3, Kevin Liu wrote: >>>>>>>>> >>>>>>>>> Tim Holy, what if I could tap into the well of knowledge that you >>>>>>>>> are to speed up things? Can you imagine if every learner had to start >>>>>>>>> without priors? >>>>>>>>> >>>>>>>>> > On Aug 9, 2016, at 07:06, Tim Holy <[email protected]> wrote: >>>>>>>>> > >>>>>>>>> > I'd recommend starting by picking a very small project. For >>>>>>>>> example, fix a bug >>>>>>>>> > or implement a small improvement in a package that you already >>>>>>>>> find useful or >>>>>>>>> > interesting. That way you'll get some guidance while making a >>>>>>>>> positive >>>>>>>>> > contribution; once you know more about julia, it will be easier >>>>>>>>> to see your >>>>>>>>> > way forward. >>>>>>>>> > >>>>>>>>> > Best, >>>>>>>>> > --Tim >>>>>>>>> > >>>>>>>>> >> On Monday, August 8, 2016 8:22:01 PM CDT Kevin Liu wrote: >>>>>>>>> >> I have no idea where to start and where to finish. Founders' >>>>>>>>> help would be >>>>>>>>> >> wonderful. >>>>>>>>> >> >>>>>>>>> >>> On Tuesday, August 9, 2016 at 12:19:26 AM UTC-3, Kevin Liu >>>>>>>>> wrote: >>>>>>>>> >>> After which I have to code Felix into Julia, a relational >>>>>>>>> optimizer for >>>>>>>>> >>> statistical inference with Tuffy < >>>>>>>>> http://i.stanford.edu/hazy/tuffy/> >>>>>>>>> >>> inside, for enterprise settings. >>>>>>>>> >>> >>>>>>>>> >>>> On Tuesday, August 9, 2016 at 12:07:32 AM UTC-3, Kevin Liu >>>>>>>>> wrote: >>>>>>>>> >>>> Can I get tips on bringing Alchemy's optimized Tuffy >>>>>>>>> >>>> <http://i.stanford.edu/hazy/tuffy/> in Java to Julia while >>>>>>>>> showing the >>>>>>>>> >>>> best of Julia? I am going for the most correct way, even if >>>>>>>>> it means >>>>>>>>> >>>> coding >>>>>>>>> >>>> Tuffy into C and Julia. >>>>>>>>> >>>> >>>>>>>>> >>>>> On Sunday, August 7, 2016 at 8:34:37 PM UTC-3, Kevin Liu >>>>>>>>> wrote: >>>>>>>>> >>>>> I'll try to build it, compare it, and show it to you guys. I >>>>>>>>> offered to >>>>>>>>> >>>>> do this as work. I am waiting to see if they will accept it. >>>>>>>>> >>>>> >>>>>>>>> >>>>>> On Sunday, August 7, 2016 at 6:15:50 PM UTC-3, Stefan >>>>>>>>> Karpinski wrote: >>>>>>>>> >>>>>> Kevin, as previously requested by Isaiah, please take this >>>>>>>>> to some >>>>>>>>> >>>>>> other forum or maybe start a blog. >>>>>>>>> >>>>>> >>>>>>>>> >>>>>>> On Sat, Aug 6, 2016 at 10:53 PM, Kevin Liu < >>>>>>>>> [email protected]> wrote: >>>>>>>>> >>>>>>> Symmetry-based learning, Domingos, 2014 >>>>>>>>> >>>>>>> >>>>>>>>> https://www.microsoft.com/en-us/research/video/symmetry-based-learning >>>>>>>>> >>>>>>>>> >>>>>>> / >>>>>>>>> >>>>>>> >>>>>>>>> >>>>>>> Approach 2: Deep symmetry networks generalize >>>>>>>>> convolutional neural >>>>>>>>> >>>>>>> networks by tying parameters and pooling over an arbitrary >>>>>>>>> symmetry >>>>>>>>> >>>>>>> group, >>>>>>>>> >>>>>>> not just the translation group. In preliminary >>>>>>>>> experiments, they >>>>>>>>> >>>>>>> outperformed convnets on a digit recognition task. >>>>>>>>> >>>>>>> >>>>>>>>> >>>>>>>> On Friday, August 5, 2016 at 4:56:45 PM UTC-3, Kevin Liu >>>>>>>>> wrote: >>>>>>>>> >>>>>>>> Minsky died of a cerebral hemorrhage at the age of >>>>>>>>> 88.[40] >>>>>>>>> >>>>>>>> <https://en.wikipedia.org/wiki/Marvin_Minsky#cite_note-40> >>>>>>>>> Ray >>>>>>>>> >>>>>>>> Kurzweil <https://en.wikipedia.org/wiki/Ray_Kurzweil> >>>>>>>>> says he was >>>>>>>>> >>>>>>>> contacted by the cryonics organization Alcor Life >>>>>>>>> Extension >>>>>>>>> >>>>>>>> Foundation >>>>>>>>> >>>>>>>> < >>>>>>>>> https://en.wikipedia.org/wiki/Alcor_Life_Extension_Foundation> >>>>>>>>> >>>>>>>> seeking >>>>>>>>> >>>>>>>> Minsky's body.[41] >>>>>>>>> >>>>>>>> < >>>>>>>>> https://en.wikipedia.org/wiki/Marvin_Minsky#cite_note-Kurzweil-41> >>>>>>>>> >>>>>>>>> >>>>>>>> Kurzweil believes that Minsky was cryonically preserved >>>>>>>>> by Alcor and >>>>>>>>> >>>>>>>> will be revived by 2045.[41] >>>>>>>>> >>>>>>>> < >>>>>>>>> https://en.wikipedia.org/wiki/Marvin_Minsky#cite_note-Kurzweil-41> >>>>>>>>> >>>>>>>>> >>>>>>>> Minsky >>>>>>>>> >>>>>>>> was a member of Alcor's Scientific Advisory Board >>>>>>>>> >>>>>>>> <https://en.wikipedia.org/wiki/Advisory_Board>.[42] >>>>>>>>> >>>>>>>> < >>>>>>>>> https://en.wikipedia.org/wiki/Marvin_Minsky#cite_note-AlcorBoard-42> >>>>>>>>> >>>>>>>>> >>>>>>>> In >>>>>>>>> >>>>>>>> keeping with their policy of protecting privacy, Alcor >>>>>>>>> will neither >>>>>>>>> >>>>>>>> confirm >>>>>>>>> >>>>>>>> nor deny that Alcor has cryonically preserved Minsky.[43] >>>>>>>>> >>>>>>>> <https://en.wikipedia.org/wiki/Marvin_Minsky#cite_note-43> >>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>>>> >>>>>>>> We better do a good job. >>>>>>>>> >>>>>>>> >>>>>>>>> >>>>>>>>> On Friday, August 5, 2016 at 4:45:42 PM UTC-3, Kevin Liu >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>> *So, I think in the next 20 years (2003), if we can get >>>>>>>>> rid of all >>>>>>>>> >>>>>>>>> of the traditional approaches to artificial >>>>>>>>> intelligence, like >>>>>>>>> >>>>>>>>> neural nets >>>>>>>>> >>>>>>>>> and genetic algorithms and rule-based systems, and just >>>>>>>>> turn our >>>>>>>>> >>>>>>>>> sights a >>>>>>>>> >>>>>>>>> little bit higher to say, can we make a system that can >>>>>>>>> use all >>>>>>>>> >>>>>>>>> those >>>>>>>>> >>>>>>>>> things for the right kind of problem? Some problems are >>>>>>>>> good for >>>>>>>>> >>>>>>>>> neural >>>>>>>>> >>>>>>>>> nets; we know that others, neural nets are hopeless on >>>>>>>>> them. Genetic >>>>>>>>> >>>>>>>>> algorithms are great for certain things; I suspect I >>>>>>>>> know what >>>>>>>>> >>>>>>>>> they're bad >>>>>>>>> >>>>>>>>> at, and I won't tell you. (Laughter)* - Minsky, founder >>>>>>>>> of CSAIL >>>>>>>>> >>>>>>>>> MIT >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *Those programmers tried to find the single best way to >>>>>>>>> represent >>>>>>>>> >>>>>>>>> knowledge - Only Logic protects us from paradox.* - >>>>>>>>> Minsky (see >>>>>>>>> >>>>>>>>> attachment from his lecture) >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>>> On Friday, August 5, 2016 at 8:12:03 AM UTC-3, Kevin >>>>>>>>> Liu wrote: >>>>>>>>> >>>>>>>>>> Markov Logic Network is being used for the continuous >>>>>>>>> development >>>>>>>>> >>>>>>>>>> of drugs to cure cancer at MIT's CanceRX < >>>>>>>>> http://cancerx.mit.edu/>, >>>>>>>>> >>>>>>>>>> on >>>>>>>>> >>>>>>>>>> DARPA's largest AI project to date, Personalized >>>>>>>>> Assistant that >>>>>>>>> >>>>>>>>>> Learns (PAL) <https://pal.sri.com/>, progenitor of >>>>>>>>> Siri. One of >>>>>>>>> >>>>>>>>>> Alchemy's largest applications to date was to learn a >>>>>>>>> semantic >>>>>>>>> >>>>>>>>>> network >>>>>>>>> >>>>>>>>>> (knowledge graph as Google calls it) from the web. >>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>>> Some on Probabilistic Inductive Logic Programming / >>>>>>>>> Probabilistic >>>>>>>>> >>>>>>>>>> Logic Programming / Statistical Relational Learning >>>>>>>>> from CSAIL >>>>>>>>> >>>>>>>>>> < >>>>>>>>> http://people.csail.mit.edu/kersting/ecmlpkdd05_pilp/pilp_ida2005_ >>>>>>>>> >>>>>>>>>> tut.pdf> (my understanding is Alchemy does PILP from >>>>>>>>> entailment, >>>>>>>>> >>>>>>>>>> proofs, and >>>>>>>>> >>>>>>>>>> interpretation) >>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>>> The MIT Probabilistic Computing Project (where there is >>>>>>>>> Picture, an >>>>>>>>> >>>>>>>>>> extension of Julia, for computer vision; Watch the >>>>>>>>> video from >>>>>>>>> >>>>>>>>>> Vikash) >>>>>>>>> >>>>>>>>>> <http://probcomp.csail.mit.edu/index.html> >>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>>> Probabilistic programming could do for Bayesian ML what >>>>>>>>> Theano has >>>>>>>>> >>>>>>>>>> done for neural networks. >>>>>>>>> >>>>>>>>>> <http://www.inference.vc/deep-learning-is-easy/> - >>>>>>>>> Ferenc Huszár >>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>>> Picture doesn't appear to be open-source, even though >>>>>>>>> its Paper is >>>>>>>>> >>>>>>>>>> available. >>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>>> I'm in the process of comparing the Picture Paper and >>>>>>>>> Alchemy code >>>>>>>>> >>>>>>>>>> and would like to have an open-source PILP from Julia >>>>>>>>> that combines >>>>>>>>> >>>>>>>>>> the >>>>>>>>> >>>>>>>>>> best of both. >>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>>> On Wednesday, August 3, 2016 at 5:01:02 PM UTC-3, >>>>>>>>> Christof Stocker >>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>>> wrote: >>>>>>>>> >>>>>>>>>>> This sounds like it could be a great contribution. I >>>>>>>>> shall keep a >>>>>>>>> >>>>>>>>>>> curious eye on your progress >>>>>>>>> >>>>>>>>>>> >>>>>>>>> >>>>>>>>>>> Am Mittwoch, 3. August 2016 21:53:54 UTC+2 schrieb >>>>>>>>> Kevin Liu: >>>>>>>>> >>>>>>>>>>>> Thanks for the advice Cristof. I am only interested >>>>>>>>> in people >>>>>>>>> >>>>>>>>>>>> wanting to code it in Julia, from R by Domingos. The >>>>>>>>> algo has >>>>>>>>> >>>>>>>>>>>> been >>>>>>>>> >>>>>>>>>>>> successfully applied in many areas, even though there >>>>>>>>> are many >>>>>>>>> >>>>>>>>>>>> other areas >>>>>>>>> >>>>>>>>>>>> remaining. >>>>>>>>> >>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>> On Wed, Aug 3, 2016 at 4:45 PM, Christof Stocker < >>>>>>>>> >>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>> [email protected]> wrote: >>>>>>>>> >>>>>>>>>>>>> Hello Kevin, >>>>>>>>> >>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>> Enthusiasm is a good thing and you should hold on to >>>>>>>>> that. But >>>>>>>>> >>>>>>>>>>>>> to save yourself some headache or disappointment >>>>>>>>> down the road I >>>>>>>>> >>>>>>>>>>>>> advice a >>>>>>>>> >>>>>>>>>>>>> level head. Nothing is really as bluntly obviously >>>>>>>>> solved as it >>>>>>>>> >>>>>>>>>>>>> may seems >>>>>>>>> >>>>>>>>>>>>> at first glance after listening to brilliant people >>>>>>>>> explain >>>>>>>>> >>>>>>>>>>>>> things. A >>>>>>>>> >>>>>>>>>>>>> physics professor of mine once told me that one of >>>>>>>>> the (he >>>>>>>>> >>>>>>>>>>>>> thinks) most >>>>>>>>> >>>>>>>>>>>>> malicious factors to his past students progress >>>>>>>>> where overstated >>>>>>>>> >>>>>>>>>>>>> results/conclusions by other researches (such as >>>>>>>>> premature >>>>>>>>> >>>>>>>>>>>>> announcements >>>>>>>>> >>>>>>>>>>>>> from CERN). I am no mathematician, but as far as I >>>>>>>>> can judge is >>>>>>>>> >>>>>>>>>>>>> the no free >>>>>>>>> >>>>>>>>>>>>> lunch theorem of pure mathematical nature and not >>>>>>>>> something >>>>>>>>> >>>>>>>>>>>>> induced >>>>>>>>> >>>>>>>>>>>>> empirically. These kind of results are not that >>>>>>>>> easily to get >>>>>>>>> >>>>>>>>>>>>> rid of. If >>>>>>>>> >>>>>>>>>>>>> someone (especially an expert) states such a theorem >>>>>>>>> will prove >>>>>>>>> >>>>>>>>>>>>> wrong I >>>>>>>>> >>>>>>>>>>>>> would be inclined to believe that he is not talking >>>>>>>>> about >>>>>>>>> >>>>>>>>>>>>> literally, but >>>>>>>>> >>>>>>>>>>>>> instead is just trying to make a point about a more >>>>>>>>> or less >>>>>>>>> >>>>>>>>>>>>> practical >>>>>>>>> >>>>>>>>>>>>> implication. >>>>>>>>> >>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>> Am Mittwoch, 3. August 2016 21:27:05 UTC+2 schrieb >>>>>>>>> Kevin Liu: >>>>>>>>> >>>>>>>>>>>>>> The Markov logic network represents a probability >>>>>>>>> distribution >>>>>>>>> >>>>>>>>>>>>>> over the states of a complex system (i.e. a cell), >>>>>>>>> comprised of >>>>>>>>> >>>>>>>>>>>>>> entities, >>>>>>>>> >>>>>>>>>>>>>> where logic formulas encode the dependencies >>>>>>>>> between them. >>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>> On Wednesday, August 3, 2016 at 4:19:09 PM UTC-3, >>>>>>>>> Kevin Liu >>>>>>>>> >>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>> wrote: >>>>>>>>> >>>>>>>>>>>>>>> Alchemy is like an inductive Turing machine, to be >>>>>>>>> programmed >>>>>>>>> >>>>>>>>>>>>>>> to learn broadly or restrictedly. >>>>>>>>> >>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>> The logic formulas from rules through which it >>>>>>>>> represents can >>>>>>>>> >>>>>>>>>>>>>>> be inconsistent, incomplete, or even incorrect-- >>>>>>>>> the learning >>>>>>>>> >>>>>>>>>>>>>>> and >>>>>>>>> >>>>>>>>>>>>>>> probabilistic reasoning will correct them. The key >>>>>>>>> point is >>>>>>>>> >>>>>>>>>>>>>>> that Alchemy >>>>>>>>> >>>>>>>>>>>>>>> doesn't have to learn from scratch, proving >>>>>>>>> Wolpert and >>>>>>>>> >>>>>>>>>>>>>>> Macready's no free >>>>>>>>> >>>>>>>>>>>>>>> lunch theorem wrong by performing well on a >>>>>>>>> variety of classes >>>>>>>>> >>>>>>>>>>>>>>> of problems, >>>>>>>>> >>>>>>>>>>>>>>> not just some. >>>>>>>>> >>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>> On Wednesday, August 3, 2016 at 4:01:15 PM UTC-3, >>>>>>>>> Kevin Liu >>>>>>>>> >>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>> wrote: >>>>>>>>> >>>>>>>>>>>>>>>> Hello Community, >>>>>>>>> >>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>> I'm in the last pages of Pedro Domingos' book, >>>>>>>>> the Master >>>>>>>>> >>>>>>>>>>>>>>>> Algo, one of two recommended by Bill Gates to >>>>>>>>> learn about AI. >>>>>>>>> >>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>> From the book, I understand all learners have to >>>>>>>>> represent, >>>>>>>>> >>>>>>>>>>>>>>>> evaluate, and optimize. There are many types of >>>>>>>>> learners that >>>>>>>>> >>>>>>>>>>>>>>>> do this. What >>>>>>>>> >>>>>>>>>>>>>>>> Domingos does is generalize these three parts, >>>>>>>>> (1) using >>>>>>>>> >>>>>>>>>>>>>>>> Markov Logic >>>>>>>>> >>>>>>>>>>>>>>>> Network to represent, (2) posterior probability >>>>>>>>> to evaluate, >>>>>>>>> >>>>>>>>>>>>>>>> and (3) >>>>>>>>> >>>>>>>>>>>>>>>> genetic search with gradient descent to optimize. >>>>>>>>> The >>>>>>>>> >>>>>>>>>>>>>>>> posterior can be >>>>>>>>> >>>>>>>>>>>>>>>> replaced for another accuracy measure when it is >>>>>>>>> easier, as >>>>>>>>> >>>>>>>>>>>>>>>> genetic search >>>>>>>>> >>>>>>>>>>>>>>>> replaced by hill climbing. Where there are 15 >>>>>>>>> popular options >>>>>>>>> >>>>>>>>>>>>>>>> for >>>>>>>>> >>>>>>>>>>>>>>>> representing, evaluating, and optimizing, >>>>>>>>> Domingos >>>>>>>>> >>>>>>>>>>>>>>>> generalized them into >>>>>>>>> >>>>>>>>>>>>>>>> three options. The idea is to have one unified >>>>>>>>> learner for >>>>>>>>> >>>>>>>>>>>>>>>> any application. >>>>>>>>> >>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>> There is code already done in R >>>>>>>>> >>>>>>>>>>>>>>>> https://alchemy.cs.washington.edu/. My question: >>>>>>>>> anybody in >>>>>>>>> >>>>>>>>>>>>>>>> the community vested in coding it into Julia? >>>>>>>>> >>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>> Thanks. Kevin >>>>>>>>> >>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>>> On Friday, June 3, 2016 at 3:44:09 PM UTC-3, >>>>>>>>> Kevin Liu wrote: >>>>>>>>> >>>>>>>>>>>>>>>>> https://github.com/tbreloff/OnlineAI.jl/issues/5 >>>>>>>>> >>>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>>> On Friday, June 3, 2016 at 11:17:28 AM UTC-3, >>>>>>>>> Kevin Liu >>>>>>>>> >>>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>>> wrote: >>>>>>>>> >>>>>>>>>>>>>>>>>> I plan to write Julia for the rest of me >>>>>>>>> life... given it >>>>>>>>> >>>>>>>>>>>>>>>>>> remains suitable. I am still reading all of >>>>>>>>> Colah's >>>>>>>>> >>>>>>>>>>>>>>>>>> material on nets. I ran >>>>>>>>> >>>>>>>>>>>>>>>>>> Mocha.jl a couple weeks ago and was very happy >>>>>>>>> to see it >>>>>>>>> >>>>>>>>>>>>>>>>>> work. Thanks for >>>>>>>>> >>>>>>>>>>>>>>>>>> jumping in and telling me about OnlineAI.jl, I >>>>>>>>> will look >>>>>>>>> >>>>>>>>>>>>>>>>>> into it once I am >>>>>>>>> >>>>>>>>>>>>>>>>>> ready. From a quick look, perhaps I could help >>>>>>>>> and learn by >>>>>>>>> >>>>>>>>>>>>>>>>>> building a very >>>>>>>>> >>>>>>>>>>>>>>>>>> clear documentation of it. Would really like to >>>>>>>>> see Julia a >>>>>>>>> >>>>>>>>>>>>>>>>>> leap ahead of >>>>>>>>> >>>>>>>>>>>>>>>>>> other languages, and plan to contribute heavily >>>>>>>>> to it, but >>>>>>>>> >>>>>>>>>>>>>>>>>> at the moment am >>>>>>>>> >>>>>>>>>>>>>>>>>> still getting introduced to CS, programming, >>>>>>>>> and nets at >>>>>>>>> >>>>>>>>>>>>>>>>>> the basic level. >>>>>>>>> >>>>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>>>> On Friday, June 3, 2016 at 10:48:15 AM UTC-3, >>>>>>>>> Tom Breloff >>>>>>>>> >>>>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>>>> wrote: >>>>>>>>> >>>>>>>>>>>>>>>>>>> Kevin: computers that program themselves is a >>>>>>>>> concept >>>>>>>>> >>>>>>>>>>>>>>>>>>> which is much closer to reality than most >>>>>>>>> would believe, >>>>>>>>> >>>>>>>>>>>>>>>>>>> but julia-users >>>>>>>>> >>>>>>>>>>>>>>>>>>> isn't really the best place for this >>>>>>>>> speculation. If >>>>>>>>> >>>>>>>>>>>>>>>>>>> you're actually >>>>>>>>> >>>>>>>>>>>>>>>>>>> interested in writing code, I'm happy to >>>>>>>>> discuss in >>>>>>>>> >>>>>>>>>>>>>>>>>>> OnlineAI.jl. I was >>>>>>>>> >>>>>>>>>>>>>>>>>>> thinking about how we might tackle code >>>>>>>>> generation using a >>>>>>>>> >>>>>>>>>>>>>>>>>>> neural framework >>>>>>>>> >>>>>>>>>>>>>>>>>>> I'm working on. >>>>>>>>> >>>>>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>>>>> On Friday, June 3, 2016, Kevin Liu < >>>>>>>>> [email protected]> >>>>>>>>> >>>>>>>>>>>>>>>>>>> >>>>>>>>> >>>>>>>>>>>>>>>>>>> wrote: >>>>>>>>> >>>>>>>>>>>>>>>>>>>> If Andrew Ng who cited Gates, and Gates who >>>>>>>>> cited >>>>>>>>> >>>>>>>>>>>>>>>>>>>> Domingos (who did not lecture at Google with >>>>>>>>> a TensorFlow >>>>>>>>> >>>>>>>>>>>>>>>>>>>> question in the >>>>>>>>> >>>>>>>>>>>>>>>>>>>> end), were unsuccessful penny traders, Julia >>>>>>>>> was a >>>>>>>>> >>>>>>>>>>>>>>>>>>>> language for web design, >>>>>>>>> >>>>>>>>>>>>>>>>>>>> and the tribes in the video didn't actually >>>>>>>>> solve >>>>>>>>> >>>>>>>>>>>>>>>>>>>> problems, perhaps this >>>>>>>>> >>>>>>>>>>>>>>>>>>>> would be a wildly off-topic, speculative >>>>>>>>> discussion. But >>>>>>>>> >>>>>>>>>>>>>>>>>>>> these statements >>>>>>>>> >>>>>>>>>>>>>>>>>>>> couldn't be farther from the truth. In fact, >>>>>>>>> if I had >>>>>>>>> >>>>>>>>>>>>>>>>>>>> known about this >>>>>>>>> >>>>>>>>>>>>>>>>>>>> video some months ago I would've understood >>>>>>>>> better on how >>>>>>>>> >>>>>>>>>>>>>>>>>>>> to solve a >>>>>>>>> & >>>>>>>> >>>>>>>>
