cbennett

I know. I had to endure the same with my sister. She was only 24- tragic bike accident. We gave her one day that way after all the tests- apnea, ear flush, even an additional EEG which we requested and was sparse as the night sky- came back negative. We looked at the weekly costs and made an informed decision as well as choosing some organ donation options.

Watching her wheeled out of that hospital room heart still beating was certainly the hardest moment of my life so far. We never saw the actual 'plug' pulled and i actually prefer(red) it that way.

I deeply empathize that the decision was all on you. doing it collectively was a hair easier. hope your emotional/personal recovery has been as swift as possible.

Glad to hear you find it exciting. I do too.. its a really hot field at the moment and I mean that in the good, not bad way ;) Lots of groups working in parallel on somewhat orthogonal design and architecture issues, with a variety of different considered devices, but a common basis set is emerging ;)

So, here's the paper I mentioned above. I think this is very methodical and inventive and definitely one of the best yet at considering confluence of DNNs and memristive (ReRAM) devices. A quick search revealed this was already on HN. https://arxiv.org/abs/1603.07341

So, I already mentioned the iconic Strukov paper above and my own which is really quite similar to Strukov in learning strategy/philosophy, except for we used entirely chemical and 'slow' devices , which may be quite interesting for brain emulation. (remember the brain operates in the mS , or microsecond regime and not nanosecond).

Here's another article I just stumbled upon a few days ago but which looks quite promising and brings us into the territory of a more un-supervised /probabilistic algorithm for learning. http://www.nature.com/articles/ncomms12611

In truth it is. To practitioners in the field building circuits or doing simulations with nanodevices, this is somewhat of a tiresome debate.

To give some background, Leon Chua made certain claims about a hypothetical fourth circuit element and these debates largely stem back to claims about circuit analysis and mathematics. Basically his models predict a perfect device which, to my knowledge , has not been experimentally realized (to the contrary of HP's claims).

However , the funny thing is it doesn't really matter. We don't need a perfect memristor to build interesting and useful nanoionic and nano-redox circuits performing non-linear computational tasks. As modelers though, we do need to be careful making ideal claims about eternal non-volatility and device life (of course). Many point out that to the contrary of being ideal, these devices are extremely variable and imperfect- which is true. Anything built using nanofab techniques at the academic level (excluding semi-con industrial processes) will be..

Btw, If you want more physics depth on this , I can recommend any paper or book by Waser. They are all good. http://eu.wiley.com/WileyCDA/WileyTitle/productCd-3527334173...

Edit: adding a link to the first book chapter of aforementioned book I found which is already rather good. https://application.wiley-vch.de/books/sample/3527334173_c01...

Overall, you are right in your diagnosis - we simply have not located a single mechanism responsible for memory. In reality, it is probably a collective effect due to a huge maelstrom of chemical computing. Our brain is one of the most imposing dynamical systems ever studied! That's why the mathematics and in turn modeling of it is so beastly.

Right now there are broadly speaking two camps in neuroscience, those who are connectionists and believe that memories are overall rather non-volatile/physically localized, with an MIT group that showed particular (individual ) neurons linked to particular memories, at the extreme of that school. On the other hand, you have neuroscientists who subscribe to a more plastic conception- in which mass synchronizations and redundant information is consistently combined and recombined.

My take is that it is no contradiction to admit the our brain is clearly volatile, and non-volatile in different modes. The transition from STP (short term memory) to LTP (long term memory) probably involves different conformational changes in neurons, dendrites, who knows even on the epigentic scale. But we do know the volatility is important and interesting-- which is precisely what makes this diffusive memristive study so important!

Short answer: yes. Basically there are two options for future ReRAM (memristive) learning system: in-situ/on-chip learning , in which all learning rules are locally derived and enforced, and ex-situ learning systems in which we do what you suggested- import weights from more computationally/power expensive substrates. there is probably abundant promise in both approaches moving forward. I recommend looking at some recent papers by the Strukov group [1] as well as my own [2] to see the limitations of these approaches. Strukov paper skirts around the issue to a certain degree but they admit in supplementary material scaling issues are not favorable with their approach. our work takes the 'neural FPGA' approach quite literally. But, their approach may , with some improvements , do rather well for an on-chip backprop implementation. Let's see what they do next. Lastly, as far as hybrid approaches, there is a recent IBM paper which is really nice which talks about deep neural net acceleration with ReRAM. If you're really curious let me know and I"ll try to dig it up. [1] http://www.nature.com/nature/journal/v521/n7550/abs/nature14... [2]http://www.nature.com/articles/srep31932

Corrected: ISIS official media wing has claimed responsibility for the attack http://www.cbsnews.com/news/brussels-attacks-isis-claims-res...

Yes, it was misleading I thought all these videos were original content of that site; in fact, they are original content produced by Welch Labs: http://www.welchlabs.com/

That's precisely right. I think between this and some of the upcoming work in memcomputing (& self-organzing electronics more broadly), we have a new set of tools which may resemble 'physical oracles' . IE, these special systems can compute exponentially difficult problems through thermodynamics or whatever other quantum-ally optimal properties biology has evolved to use. They may not be universal computers at first, but they ultimately may point the way forward towards that vision.

thank you for linking it through.

The pdf is pay-walled on Nature; thankfully, it is up on Arxiv too: http://arxiv.org/pdf/1507.08993v1.pdf

Upset about this issue and want to make a positive difference rather than debating technicalities or simply waiting for the other shoe to drop?

It may be no surprise that FATCA and FBAR present several consitutitional issues. Chief among them are the following:

-As the IRS can unilaterally sign unauthorized pseudo-treaties with foreign governments, the Senate’s role in ratifying treaties is bypassed

-Fourth amendment: Private financial data is collected indiscriminately

-Eighth Amendment prohibiting cruel and unusual punishment, as well as excessive fines, may be leveraged when compliance is not perfect, and without Rule of Law provisions to ensure any protection for the accused. (My recapitulation of points in [1] )

Thankfully, several large scale legal challenges on exactly these grounds are now underway- by Sen Rand Paul and 6 other expats [2] , and perhaps even more promisingly, by super lawyer Jim Bopp (who took down Mc-Cain Feingold [3]).

You can also donate a small amount to support the latter case: https://fatcalegalaction.com/contribute/ Thanks for reading, lets not give up hope, and remain fired up to the correctness of this cause because, in the words of Bopp - “The U.S. Constitution protects every citizen’s liberty and freedom, while FATCA undermines both,”

We deserve better!

Signed, An American living in Paris

[1] http://www.thenewamerican.com/usnews/constitution/item/18253...

[2] http://www.taxanalysts.com/www/features.nsf/Features/9D07FB4...

[3] http://www.washingtontimes.com/news/2014/may/5/superlawyer-j...

great read. thanks for that.

For a dramatic international comparison, consider that the 15th-20th largest cities in China already have populations of greater than 7M: double this forecast to be reached over the next decade in America's 3rd Largest urban area

https://en.wikipedia.org/wiki/List_of_urban_agglomerations_i...

"An interesting report, but is there any reason to conclude that "we are due"?"

Yes, El Nino, which may bring torrents of rain this winter to the US west coast. The combination with epic drought will be disastrous, whether or not if fulfills the mythic criterion of an "ArkStorm'

FYI: occurrence of rain or lack thereof is not like flipping a coin; this is more like a chaotic system with non-linear patterns evolving; once we see a large-scale pattern that occurs, and we identify it recurring through time such as El Nino, we can spot its tell-tale signs ahead of time and warn folks that worse may be on the way. Nevertheless, there are no guarantees, all we can do is wait and watch- But it certainly changes our Prior to know about these recurrent global-weather-scale patterns.

http://www.aljazeera.com/news/2015/09/el-nino-set-strongest-...

TLDR: There is a major mismatch between the device physics of the academic group and what is being marketed/sold.

If the actual device they were selling represented what they did in this technical work, I would be very, very excited. The I/V curve looks great.

BUT, compare that figure to the I/V curve represented in the longer datasheet (linked to in parent comment). A one million fold increase in resistance in 5nS- That isn't analog at all. It could be fairly emulated as almost a flip-flop with just a few transistors!! Although a binary memristor at scale might be more interesting as a memory (eg ReRAM), at this scale meant for neuromorphic analog applications, it becomes utterly uninteresting.

So to be honest I am baffled how they went from such a promising inspiration to a completely bland packaged device. Two possibiliies come to mind: 'they' are entirely different groups, or a major mis-marketing is occurring in the interest of gaining some funds for future R&D.

Its not just for the AI field, if you read their About Us page you will see that a very interesting application has to do with UAV and algorithms for motion detection. The chip set design space for this is very low energy but efficient/accurate at a certain trained task, which is what makes bio-inspired applications so compelling. But going back to AI, I would say this:

Short term: a wide variety of demonstrators of various 'easy' problems, eg image and simple logic function pattern matching, in chip as opposed to in theory or software like you say. This can be done with dozens or a few hundred memristive devices.

Medium term: more interesting (my maybe not trying NP hard problems yet) application in image, sound processing and other non-linear signal processing that begin to rival best approaches coming from dozens of GPU cores. Such applications might require tens of thousands or even hundreds of thousands of devices. Pattern matching and classifications at this stage might rival that of a very simple brain or a small component of the neo-cortex.

Longer term: Unknown, but some possibilities when you use tens of millions of such devices may include full (neo)-cortex implementation in hardware, remember that an entirely new generation of non-linear algorithms that implement due to topological and temporal effects at the nano-scale may allow for better results than what we get given an equivalent number of transistors (yes, transistors are also nano-devices, but despite a small community working on sub-threshold switching, they are being used in an almost uniformly binary approach).

Related course readings and assignments for self-learning: http://ocw.mit.edu/courses/mathematics/18-s996-category-theo... couldn't manage to find any videos, though :/

thanks for that.

I just googled 'Improbable London'. On their website not much substantive , just a standard 'hello world' post: http://improbable.io/blog/

I'm personally interested in such an approach for an academic tasks ( complex electrical simulations), but I have no idea where to find some substantive discussion of what they're doing here.

kind of sad

vapid article but more than that the UI for actually getting through the story is awful. the recommended articles box above the much smaller 'Continue' button, is particularly egregious.

Having lived in Belgium, Sweden, and now in France - 3 different EU countries in 3 years- I can 100% confirm the parent commenter's almost comical ignorance about the heterogeneity of European life and mores.

The paper is submitted but not accepted (passed review yet), in this you're right, but I think its a growing practice for reputable scientists (as well as nobodies) to post to Arxiv immediately after they've submitted to a 'top' journal.

Anyways, here's the lead authors page, for the curious: http://physics.ucsd.edu/~diventra/

Thanks for linking to this, Aaronson brings some good critiques (as did others on this thread). I do however think it is a bit disrespectful to the authors to use the word debunk,. Debunk implies a myth or a belief, or a completely unsupported theory, such as creationism. In contradistinction, this is peer reviewed science, and while there are undoubtedly probably faulty assumptions behind their work (or merely overstated significance), there may be some underlying validity and modest importance within the field of unconventional computing as well, so let us not throw out the baby with the bathwater.

Also, just curious as a parallel example, Tononi and co had their day in the sun on Aaronson's blog a few months back, and there was a vigorous back and forth including a great response from Tononi ; would you say that now IIT has been 'debunked'? If so, how do you justify that? If not, why not wait for the authors of UMM theory and its hardware extension to have a chance to reply? This is how progress is made in science, deliberation & iteration, not instant victories.

"Also they are not ambitious enough. Assuming constant time arithmetic of the real numbers should get you more (like maybe even the halting problem)."

Agreed. They do at least broach the idea though, this is from the companion paper (more theoretical):

"We have thus shown that a UMM is Turing-complete, namely it can simulate any Turing machine (whether deterministic or not). Note, however, that the reverse is not necessarily true. Namely, we have not demonstrated that a UTM can simulate a UMM, or, equivalently, we have not demonstrated that a UMM is Turing-equivalent. It is worth pointing out that, if we could prove that a UMM is not Turing-equivalent, some (Turing) undecidable problems, such as the halting problem [2], may find solution within our UMM paradigm, thus contradicting the Church-Turing hypothesis [2]. Although this is an intriguing–albeit unlikely– possibility, we leave its study for future work." http://arxiv.org/pdf/1405.0931.pdf

thanks very much for this! had been looking for some time for a quite recent Williams talk

Here is the paper for any one curious in a look at the actual configuration of the nanowire device: thankfully there is no pay-gating http://pubs.acs.org/doi/abs/10.1021/nn505139m

“I doubt mr Obama is at all African”

<Baracks> father, Barack Obama, Sr., was a Luo from Nyang’oma Kogelo, Kenya

http://en.wikipedia.org/wiki/Barack_Obama

Actually, in contradistinction to buzz-wordiness, most of the terms he used such as tensegrity have quite precise definitions- conceptually, as well in mechanical terms. If you want to get a true introduction to the depth and breadth of his thinking, I highly recommend his magnum opus 'Synergetics' [1]. Otherwise, at least take a glance at the meta-structure/heuristics behind his geometrical and design work [2].

[1]http://www.amazon.com/Synergetics-Explorations-Geometry-Buck...

[2]http://en.wikipedia.org/wiki/Synergetics_(Fuller)

You are right to be skeptical in the fact that anecdotal evidence from a yachting voyage could establish a general case for mass bio-mass (or lets just say a sub-set of bio-mass, like catchable fish) decline, and your point that large portions of the Pacific are normally dead-zones is true because of the way that anoxic regions can develop due to oceanic circulations of nutrients and dissolved oxygen concentrations.

But, you might be concerned to learn that the NOAA typically only reports on mass die offs fairly close to US shores, at least within our EEZ zone- because they are only adequately funded to do so- [1] is an example of such a report. There is no well-funded transnational environmental authority that can make a reliable 'alarm' about the potential of mass common-water oceanic die-offs (think 'tragedy of the commons'); Greenpeace is more of an advocacy organization than a scientific or monitoring authority.

One thing we can indirectly empirically gauge is the size of these large garbage patches at sea simply via imaging. In fact NOAA has put together a fact sheet on this [2], although it is rather simplistic. What we can glean from this is the fact that they are real, quite large, and if you look at the map it looks as though the OP's voyage passed through it at least once if not twice. Which would explain much of what he saw, but not necessarily the consistency (pervasiveness) of the dead zones observed..

Finally, just food for thought (er.. perhaps bad choice of words.. haha) that there is a sort of conflation in the article about the long-going decline of fish stocks and accumulation of garbage due to dumping and the more acute poisoning/death of fish stocks due to radioactive leakage from Fukushima in the areas off-coast Japan specifically. There is academic evidence [3] that large fish, specifically tuna, are acting as vector to transport large amounts of spoiled biomass as far afield as California. How much of the biomass die-off does this explain? I have absolutely no idea, but since both issues are quite concerning in their own right we should probably keep tabs on both..

[1] http://www.dailykos.com/story/2013/07/30/1227489/-Mass-Aquat...

[2] http://marinedebris.noaa.gov/info/pdf/patch.pdf

[3] http://www.pnas.org/content/109/24/9483.short

If definitely already does, and I think there is a systematic under-appreciation of just how catastrophic Fukushima could become if there is an additional criticality if not a series of them, it is truely an order of magnitude bigger of a concern than Chernobly ever was. The upper band for relocation might be in the tens of millions instead of hundreds of thousands:

"a newly stated concern is the proximity of melted fuel in relation to the Tokyo aquifer that extends under the plant. If and when the corium reaches the Tokyo aquifer, serious and expedient discussions will have to take place about evacuating 40 million people from the greater metropolitan area"

Of course this is a 'worst case' but given that the likelihood of a perfect mitigation is so extremely rare planners need to be at least considering this at a non-zero probability, which is more than a bit frightening.

More here:

http://rt.com/news/fukushima-apocalypse-fuel-removal-598/

And the bad news gets worse, this is from yesterday:

"Officials of the Tokyo Electric Power Company (TEPCO) said on Friday they detected 400,000 becquerels per liter of beta ray-emitting radioactive substances - including strontium - at the site, a level 6,500 times higher than readings taken on Wednesday, NHK World reported."

"On Wednesday TEPCO said it had detected high levels of radiation in a ditch leading to the Pacific Ocean, and that it suspected heavy rains had lifted contaminated soil."

We are talking about some non-zero (lets just say integer) chance of global catastrophe if the oxgenation of the Pacific ocean is disrupted by mass radioactive spoiling of biomass. I don't understand why a concerted organisation of nuclear engineers and assistants are not stepping in to avert what could be game over for Japan in general, to say nothing of cascading affects across the oceans and N. Hemisphere in general.

http://rt.com/news/fukushima-high-radioactivity-well-335/