Ethan and Jeff, Thanks for the suggestions! Excited to check them out.
- Chetan On Thu, Oct 17, 2013 at 2:51 PM, Jeff Hawkins <[email protected]> wrote: > Ethan, > This is an excellent list! > > My hesitation in answering Chetan's question is it is hard to find papers > that are "relatively straight forward". By this I assume you mean papers > that are readable by a non-neuroscientist. Even the most basic papers > assume you know a lot of jargon. But Ethan's list is an excellent start. > > I also would recommend Vernon Mountcastle's "Perceptual Neuroscience: The > Cerebral Cortex". It is a beautiful book and well written. It will give > you a good overview of the cortex but not a clue as to how it works. > > Bartlett Mel has written the most important papers on the local dendritic > properties we use in the CLA. These are technical. > > The book I mentioned about consciousness is "Why Red Doesn't Sound Like a > Bell" by Kevin O'Regan. This paper > http://cogsci.uni-osnabrueck.de/~nbp/PDFs_Publications/ORegan.BBS.01pdf.pdf > presents the same idea as the book and I think is easier to read. His > thesis is applied to what he calls "visual consciousness" but I don't know > why he restricts it to vision. > > Jeff > > > Date: Wed, 16 Oct 2013 20:30:24 -0400 > From: Ethan Berl <[email protected]> > To: "NuPIC general mailing list." <[email protected]> > Subject: Re: [nupic-dev] Neuroscience / AI papers > Message-ID: > < > capog6oiuwaacjmmyi343hegfbsdsc+70msrm6ggiobgjh2s...@mail.gmail.com> > Content-Type: text/plain; charset="windows-1252" > > I would start with the inspiring papers listed in the back of the book On > Intelligence. I am also working through these which is why I knew about > them. Best of luck! > > Mountcastle, Vernon B. "An Organizing Principle for Cerebral Function: The > Unit Model and the Distributed System," in Gerald M. Edelman and Vernon B. > Mountcastle, eds., The Mindful Brain (Cambridge, Mass.: MIT Press, 1978). > Creutzfeldt, Otto D. "Generality of the Functional Structure of the > Neocortex," Naturwissenschaften, vol. 64 (1977): pp. 507?17. > Felleman, D. J., and D. C. Van Essen. "Distributed Hierarchical Processing > in the Primate Cerebral Cortex," Cerebral Cortex, vol. 1 (January/February > 1991): pp. 1?47. > Sherman, S.M., and R.W. Guillery. "The Role of the Thalamus in the Flow of > Information to the Cortex," Philosophical Transactions of the Royal Society > of London, vol. 357, no. 1428 (2002): pp. 1695?708. > Rao, R. P., and D. H. Ballard. "Predictive Coding in the Visual Cortex: A > Functional Interpretation of Some Extra-Classical Receptive-field Effects," > Nature Neuroscience, vol. 2, no. 1 (1999): pp. 79?87. > Guillery, R. W. "Branching Thalamic Afferents Link Action and Perception," > Journal of Neurophysiology, vol. 90 (2003): pp. 539?48. > Young, 170 M. P. "The Organization of Neural Systems in the Primate > Cerebral > Cortex," Proceedings of the Royal Society: Biological Sciences, vol. 252 > (1993): pp. 13?18. > > > > > On Wed, Oct 16, 2013 at 7:35 PM, Chetan Surpur <[email protected]> wrote: > > > Hi Jeff, > > > > Glad to see you're back. Hope your travels went well. > > > > I recently watched an interview you gave on your neuroscience / AI > > journey (http://www.youtube.com/watch?v=e5xyF84dw2o). You mentioned > > that you spent a lot of time finding and reading research papers in > > your time as a grad student. > > > > Since you must have sifted through a whole lot, looking for highest > > quality ones, I wanted to ask: do you have a list of papers for > > beginners in the field to get started with? Preferably ones that are > > relatively straightforward to read, and pack a big punch. It would > > really help interested people like me hit the ground running. Thanks in > advance! > > > > Also, in your talk at NASA Ames, you mentioned a good book about > > consciousness, and asked us to email this list to get the name from you. > > Please do share if you remember it! > > > > Thanks, > > Chetan > > > > _______________________________________________ > > nupic mailing list > > [email protected] > > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > > > > > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > < > http://lists.numenta.org/pipermail/nupic_lists.numenta.org/attachments/2013 > 1016/5f0ac3ea/attachment-0001.html> > > ------------------------------ > > Message: 3 > Date: Wed, 16 Oct 2013 18:03:06 -0700 > From: Danese Cooper <[email protected]> > To: "NuPIC general mailing list." <[email protected]> > Cc: "NuPIC general mailing list." <[email protected]> > Subject: Re: [nupic-dev] Proposal: Switch from JIRA to Github Issue > Tracker? > Message-ID: <[email protected]> > Content-Type: text/plain; charset=us-ascii > > I think it's a great idea to open up this decision to the NuPIC Community. > I > am in favor of whatever most community devs prefer, so long as Numenta > engineers will use it publicly and on an even footing with the Community. I > would not be in favor of a switch if it created a "two-class system" in any > way. It looks from Matt's second message in this thread that switching > would > depend on some work to automate the flow of information from Numenta's > internal JIRA instance to NuPIC's GitHub tracker as appropriate. I'm glad > to > see that acknowledgement of integration work to be done. > > Matt, is that work you'd expect to see Numenta doing, or are you hoping to > organize the Community to pick it up? > > Danese > > On Oct 16, 2013, at 4:21 PM, Tim Boudreau <[email protected]> wrote: > > > +1 to ditching JIRA - Github has done a very nice job of making their > issue tracker as simple as it needs to be (kind of an object lesson in how > removing features can make software more powerful). Whereas JIRA's UI is > click-heavy, clunky and goshawful - classic software you sell to management > which makes developers miserable. > > > > -Tim > > > > _______________________________________________ > > nupic mailing list > > [email protected] > > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > > > > ------------------------------ > > Message: 4 > Date: Thu, 17 Oct 2013 09:37:18 +0800 > From: Stewart Mackenzie <[email protected]> > To: "NuPIC general mailing list." <[email protected]> > Subject: Re: [nupic-dev] Proposal: Switch from JIRA to Github Issue > Tracker? > Message-ID: <[email protected]> > Content-Type: text/plain; charset="utf-8" > > I am in full support of scrapping Jira, contributions will go much more > steadily. > > Only code with a clear issue detailing the clear problem gets merged into > mainline. This must be law that goes hand in hand with numenta's engineers > on equal footing regarding information exposure to issues. > > Kind regards > > Stewart > > Matthew Taylor <[email protected]> wrote: > >This is a proposal (based on our "lazy consensus" model[1]) to migrate > >our issue tracking system from JIRA[2] into Github Issues[3]. > > > >We initially choose JIRA in order to make collaboration between > >internal Grok products (using JIRA) and the NuPIC project easier. It > >provided the same workflow for our planning meetings, scrums, etc. At > >this point, I don't believe the benefits we've received by doing this > >outweigh the simplicity and automatic integration of using Github > >Issues. > > > >I'd like to open this topic up for discussion to anyone with an > >opinion. Here are my thoughts, which are admittedly biased towards > >migrating to Github. I'd especially like to hear any dissenting > >opinions. > > > >1. JIRA requires users to create yet another user account to take part > >in ticket tracking 2. JIRA / Github integration is not as easy as GH > >issue integration 3. PRs are automatically tracked with GH issues, but > >usually go untracked in JIRA (unless they are manually linked to a JIRA > >ticket, which doesn't usually happen). > >4. JIRA does give us a nice "Agile View[4]" of our sprints, but I've > >found a project free for open source called huboard[5] that gives us > >the same functionality. > >5. With only GH and no JIRA, the identities of those commenting on > >tickets and committing code is obvious, while with JIRA I have no clue > >in some cases who people are, if they have signed our license, or if > >they have ever committed code. > >6. JIRA is complicated and flexible, and it takes discipline for all > >engineers on a team to use it in the same fashion. I cannot force > >everyone in the community to use all the standards necessary for a > >"rich JIRA experience". JIRA works great for Grok internally, but we > >all know (and have fought with) our own JIRA standards, and we have a > >PM who enforces the rules. I can't realistically enforce the same level > >of constraints on how JIRA tickets are created with the OS community, > >and I don't have time to clean up everyone's tickets. > >7. GH Issues would make my job easier, because I have one less system > >to deal with :) > > > >I want to move on this rather quickly, so please respond with your > >opinions by the end of this week. I'm going to be experimenting with GH > >issues on my own, and testing ticket migration scripts in the meantime. > > > >I will assume that silence == agreement. ;) > > > >Thanks! > >--------- > >Matt Taylor > >OS Community Flag-Bearer > >Numenta > > > >[1] > >https://github.com/numenta/nupic/wiki/Contributor-Model#51-lazy-consens > >us > >[2] http://issues.numenta.org > >[3] https://github.com/features/projects/issues > >[4] https://issues.numenta.org/secure/RapidBoard.jspa?rapidView=2 > >[5] https://huboard.com/ > > > >_______________________________________________ > >nupic mailing list > >[email protected] > >http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > > -- > Sent from my Android device with K-9 Mail. Please excuse my brevity. > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > < > http://lists.numenta.org/pipermail/nupic_lists.numenta.org/attachments/2013 > 1017/228ab99d/attachment-0001.html> > > ------------------------------ > > Message: 5 > Date: Wed, 16 Oct 2013 23:31:17 -0300 > From: Pedro Tabacof <[email protected]> > To: "NuPIC general mailing list." <[email protected]> > Subject: [nupic-dev] Kaggle team assemble > Message-ID: > <CAAkoWuBQdib1TX-=wNBnD040LUp3Jt0_Bj= > [email protected]> > Content-Type: text/plain; charset="iso-8859-1" > > Hello, > > After the warm reception of the Kaggle competition idea, I decided to > create > the thread asap. For those who are not aware, Kaggle is a data mining > competition website. It has been hyped on the machine learning media and > has > some great people participating in it. It rewards with different prizes, > but > my focus is not earn money, but rather to see how good NuPIC would fare > against the state of the art. I found those following competitions to be a > viable start: > > 1) Accelerometer Biometric Competition > >From real acceleration data you have to guess from which cell phone > >came > the test samples. > 36 days to go > Pros: > -Multivariable, temporal data seems ideal for the CLA > Cons: > -The best competitors are using data leaks (meta-information), so the > chance > of winning fairly is zero (though they will give a consolation prize to the > best "honest" entry) > Questions: > -How fast would NuPIC go through a 1GB CSV file? > -The test data is not open-ended, that is, you have to say whether a > recording belongs to a specific cell phone or not, with 50% chance of being > right. Would NuPIC do well in this case? > > 2) Multi-Label Bird Species Classification Try to guess the bird species > present on some audio recordings > 38 days to go > Pros: > -Temporal data > Cons: > -Multiple labels, so I think we would need 87 different models, which is > kinda impractical > Question: > -Is there a better way to do multi-label classification with NuPIC? > > 3) Conway's Reverse Game of Life > Reverse the game of life for 1 to 5 steps > 4 months to go > Pros: > -Natural input sparseness (have to check on it) -Temporal data -Multistep > classification -We can generate our own training data if necessary > Cons: > -Maybe there are very straightforward ways to solve this (rule based, brute > force, etc) -Too much time before the competition ends > Questions: > -How sparse does the input need to be for the CLA to work well? > -Is it possible to use a 2D arrangement on the cells connection matrix? > > There is also the AMS 2013-2014 Solar Energy Prediction contest, but I > didn't have time to go over it. The other competitions don't seem to be > suited for NuPIC. > > Who here is interested on participating and what competition do you think > to > be the most promising one? > > Pedro. > -- > Pedro Tabacof, > Unicamp - Eng. de Computa??o 08. > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > < > http://lists.numenta.org/pipermail/nupic_lists.numenta.org/attachments/2013 > 1016/cb712ace/attachment-0001.html> > > ------------------------------ > > Message: 6 > Date: Wed, 16 Oct 2013 20:03:15 -0700 > From: Chetan Surpur <[email protected]> > To: "NuPIC general mailing list." <[email protected]> > Subject: Re: [nupic-dev] Kaggle team assemble > Message-ID: > <CAD1_cr=b2LywOyBiroNKERsB+h5U_F= > [email protected]> > Content-Type: text/plain; charset="iso-8859-1" > > I'm interested in participating! > > I haven't decided on one yet, but I wanted to point out that the CLA looks > ideally suited for the AMS 2013-2014 Solar Energy Prediction contest. From > what I can tell, it's about predicting the daily energy output of solar > farms from 12, 15, 18, 21, 24-hourly training data. There's 29 days to go. > > > On Wed, Oct 16, 2013 at 7:31 PM, Pedro Tabacof <[email protected]> wrote: > > > Hello, > > > > After the warm reception of the Kaggle competition idea, I decided to > > create the thread asap. For those who are not aware, Kaggle is a data > > mining competition website. It has been hyped on the machine learning > > media and has some great people participating in it. It rewards with > > different prizes, but my focus is not earn money, but rather to see > > how good NuPIC would fare against the state of the art. I found those > > following competitions to be a viable start: > > > > 1) Accelerometer Biometric Competition From real acceleration data you > > have to guess from which cell phone came the test samples. > > 36 days to go > > Pros: > > -Multivariable, temporal data seems ideal for the CLA > > Cons: > > -The best competitors are using data leaks (meta-information), so the > > chance of winning fairly is zero (though they will give a consolation > > prize to the best "honest" entry) > > Questions: > > -How fast would NuPIC go through a 1GB CSV file? > > -The test data is not open-ended, that is, you have to say whether a > > recording belongs to a specific cell phone or not, with 50% chance of > > being right. Would NuPIC do well in this case? > > > > 2) Multi-Label Bird Species Classification Try to guess the bird > > species present on some audio recordings > > 38 days to go > > Pros: > > -Temporal data > > Cons: > > -Multiple labels, so I think we would need 87 different models, which > > is kinda impractical > > Question: > > -Is there a better way to do multi-label classification with NuPIC? > > > > 3) Conway's Reverse Game of Life > > Reverse the game of life for 1 to 5 steps > > 4 months to go > > Pros: > > -Natural input sparseness (have to check on it) -Temporal data > > -Multistep classification -We can generate our own training data if > > necessary > > Cons: > > -Maybe there are very straightforward ways to solve this (rule based, > > brute force, etc) -Too much time before the competition ends > > Questions: > > -How sparse does the input need to be for the CLA to work well? > > -Is it possible to use a 2D arrangement on the cells connection matrix? > > > > There is also the AMS 2013-2014 Solar Energy Prediction contest, but I > > didn't have time to go over it. The other competitions don't seem to > > be suited for NuPIC. > > > > Who here is interested on participating and what competition do you > > think to be the most promising one? > > > > Pedro. > > -- > > Pedro Tabacof, > > Unicamp - Eng. de Computa??o 08. > > > > _______________________________________________ > > nupic mailing list > > [email protected] > > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > > > > > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > < > http://lists.numenta.org/pipermail/nupic_lists.numenta.org/attachments/2013 > 1016/e6edc232/attachment-0001.html> > > ------------------------------ > > Message: 7 > Date: Wed, 16 Oct 2013 21:07:40 -0700 > From: Matthew Taylor <[email protected]> > To: "NuPIC general mailing list." <[email protected]> > Subject: Re: [nupic-dev] Kaggle team assemble > Message-ID: <[email protected]> > Content-Type: text/plain; charset="utf-8" > > These also sound like good hackathon projects! > > Matt > > Sent from my MegaPhone > > > On Oct 16, 2013, at 8:03 PM, Chetan Surpur <[email protected]> wrote: > > > > I'm interested in participating! > > > > I haven't decided on one yet, but I wanted to point out that the CLA > looks > ideally suited for the AMS 2013-2014 Solar Energy Prediction contest. From > what I can tell, it's about predicting the daily energy output of solar > farms from 12, 15, 18, 21, 24-hourly training data. There's 29 days to go. > > > > > >> On Wed, Oct 16, 2013 at 7:31 PM, Pedro Tabacof <[email protected]> > wrote: > >> Hello, > >> > >> After the warm reception of the Kaggle competition idea, I decided to > create the thread asap. For those who are not aware, Kaggle is a data > mining > competition website. It has been hyped on the machine learning media and > has > some great people participating in it. It rewards with different prizes, > but > my focus is not earn money, but rather to see how good NuPIC would fare > against the state of the art. I found those following competitions to be a > viable start: > >> > >> 1) Accelerometer Biometric Competition From real acceleration data > >> you have to guess from which cell phone came the test samples. > >> 36 days to go > >> Pros: > >> -Multivariable, temporal data seems ideal for the CLA > >> Cons: > >> -The best competitors are using data leaks (meta-information), so the > >> chance of winning fairly is zero (though they will give a consolation > >> prize to the best "honest" entry) > >> Questions: > >> -How fast would NuPIC go through a 1GB CSV file? > >> -The test data is not open-ended, that is, you have to say whether a > recording belongs to a specific cell phone or not, with 50% chance of being > right. Would NuPIC do well in this case? > >> > >> 2) Multi-Label Bird Species Classification Try to guess the bird > >> species present on some audio recordings > >> 38 days to go > >> Pros: > >> -Temporal data > >> Cons: > >> -Multiple labels, so I think we would need 87 different models, which > >> is kinda impractical > >> Question: > >> -Is there a better way to do multi-label classification with NuPIC? > >> > >> 3) Conway's Reverse Game of Life > >> Reverse the game of life for 1 to 5 steps > >> 4 months to go > >> Pros: > >> -Natural input sparseness (have to check on it) -Temporal data > >> -Multistep classification -We can generate our own training data if > >> necessary > >> Cons: > >> -Maybe there are very straightforward ways to solve this (rule based, > >> brute force, etc) -Too much time before the competition ends > >> Questions: > >> -How sparse does the input need to be for the CLA to work well? > >> -Is it possible to use a 2D arrangement on the cells connection matrix? > >> > >> There is also the AMS 2013-2014 Solar Energy Prediction contest, but I > didn't have time to go over it. The other competitions don't seem to be > suited for NuPIC. > >> > >> Who here is interested on participating and what competition do you > think > to be the most promising one? > >> > >> Pedro. > >> -- > >> Pedro Tabacof, > >> Unicamp - Eng. de Computa??o 08. > >> > >> _______________________________________________ > >> nupic mailing list > >> [email protected] > >> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > > > > _______________________________________________ > > nupic mailing list > > [email protected] > > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: > < > http://lists.numenta.org/pipermail/nupic_lists.numenta.org/attachments/2013 > 1016/4785b9c0/attachment-0001.html> > > ------------------------------ > > Message: 8 > Date: Thu, 17 Oct 2013 08:16:46 -0700 > From: Matthew Taylor <[email protected]> > To: "NuPIC general mailing list." <[email protected]> > Subject: Re: [nupic-dev] Proposal: Switch from JIRA to Github Issue > Tracker? > Message-ID: > < > cajv6ndpspyzisdvt+vo5pgomsr-r+ubxcg4x-+f448w7dkr...@mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > On Wed, Oct 16, 2013 at 6:03 PM, Danese Cooper <[email protected]> wrote: > > It looks from Matt's second message in this thread that switching would > depend on some work to automate the flow of information from Numenta's > internal JIRA instance to NuPIC's GitHub tracker as appropriate. I'm glad > to > see that acknowledgement of integration work to be done. > > > > Matt, is that work you'd expect to see Numenta doing, or are you hoping > to > organize the Community to pick it up? > > Any automated integration is going to be dirty. At this point, I think it > makes sense to do it manually over time. There are only 152 unresolved > issues, which is manageable. > > It would help if the community helped move over tickets that are assigned > to > them by creating the new issues in Github (but not yet! > I'm still figuring how exactly how I want to use GH Issues). I will ask > for > some assistance once I have some issues moved over for the next couple > sprints, and I've established the labels and milestones we'll be using in > the new system. > > We'll keep JIRA around and reference the old tickets from the new ones, > because there are some good comments on some of the JIRA items that won't > be > easy to move over. > > --------- > Matt Taylor > OS Community Flag-Bearer > Numenta > > > > ------------------------------ > > Message: 9 > Date: Thu, 17 Oct 2013 08:17:36 -0700 > From: Matthew Taylor <[email protected]> > To: "NuPIC general mailing list." <[email protected]> > Subject: Re: [nupic-dev] Proposal: Switch from JIRA to Github Issue > Tracker? > Message-ID: > <CAJv6nDOOY= > [email protected]> > Content-Type: text/plain; charset=ISO-8859-1 > > On Wed, Oct 16, 2013 at 6:37 PM, Stewart Mackenzie <[email protected]> > wrote: > > I am in full support of scrapping Jira > > I knew I could count on you, Stewart! :) > > --------- > Matt Taylor > OS Community Flag-Bearer > Numenta > > > > ------------------------------ > > Subject: Digest Footer > > _______________________________________________ > nupic mailing list > [email protected] > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > > > ------------------------------ > > End of nupic Digest, Vol 6, Issue 22 > ************************************ > > > _______________________________________________ > nupic mailing list > [email protected] > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >
_______________________________________________ nupic mailing list [email protected] http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
