Category Archives: Unconventional Computing

Rethinking Computing

This month’s IEEE Computer Magazine has several articles on “rethinking computers”. (This year is the fiftieth anniversary of the IEEE Computer Society, with which has been marked with plenty of looking back.)

In an interesting article Christof Teuscher considers some “blank slate” ideas for totally new approaches to computing.

He sketches some familiar problems for conventional technology, heat dissipation, energy consumption, physical size limits on transistors, and limits on clock speed. He also notes the prohibitive costs of further improvements.

Teuscher is a hardware guy, so he leaves off the fact that we barely know how to program the chips we have, and such software we do have is laughably insecure and unreliable.

So, as we are now trying to reboot computing , to find innovations that go beyond incremental improvements?

Teuscher suggests one source is to look at older ideas that have already gone through part of the “hype cycle”, such as neural networks. From high hopes and unreasonable expectations through disappointments, neural networks have arrived at “the plateau of productivity”. Actually, they are ubiquitous, though always in hybrid systems with mixed architectures.

unconventional-computing technology seems to have trouble cross-ng the Trough of Disillusionment and reaching the Plateau of Productivity. “ (p. 54)

He proposes a catechism for proposed new technologies.

  • What challenge (or problem or application) are you trying to address with an unconventional- computing approach?
  • What are the metrics for meet- ing that challenge?
  • How is the system controlled and programmed?
  • What fundamental limits to computing should you be concerned about?
  • How do you interface with your unconventional system?

This is as good of a list as I have ever seen.

What is this “Unconventional Computing” Teuscher looks for “The Weird, the SMall, The Uncrollable.“  He calls this “intrinsic” computation.

By using the general concept of intrinsic computation,we can harness a substrate’s intrinsic dynamics to perform use- ful information processing—that is, solve a given computational task..” (p. 57 )

One general point is that CMOS chip design is top down, and dedicated to difficult (and expensive) manipulations of Silicon to make what we want. Bottom up approaches take the natural behavior and properties of a system to create a device. “the lack of control over bottom-up self-assembly processes often leads to designs we cannot fully control and understand.” (p. 55)

He describes the example “Reservoir Computing”, which was proposed by Alan Turing. These systems are bottom up and uncontrolled, but he says that the idea is actually quite reasonable for many new nanomaterials and processes.

Chemical and biochemical systems, including DNA, are another “Unconventional” approach. These systems are small and fast, and are programmed via learning. Scaling and hierarchical composition are still challenges to be solved.

Teuscher summarizes these examples as demonstrating how these are different from conventional computing machines.

First, to perform useful computation, we do not have to be in complete control of the physical substrate. Quite the contrary.

And second, “many nontraditional computing approaches are highly specialized. “ There will not be a general purpose CPU, but a hybrid of specific devices.

The latter point is certainly “weird” to conventional techies, and, as he points out, means that these new technologies will depend on the ability to create and train new devices to solve each newproblem. This will be a great frontier for software and CAD systems, to say the least.

Cool.


  1. Christof Teuscher, The Weird, the Small, and the Uncontrollable: Redefining the Frontiers of Computing. Computer, 50 (8):52-58, 2017.

Chemical Channels For Digital Communication

Vahid Jamali and colleagues at Stanford are exploring digital communication via chemicals. To be sure, biological organisms signal through chemical channels, including, notably neural communication. And humans have coopted many chemical messaging systems to do our own bidding.  (E.g., most drugs work by fiddling with natural chemical communications.)

However, this project is using chemical signals to transmit arbitrary digital bits, which can encode whatever messages we want.

Setting aside the question “why?” for a moment, we note that this is pretty “out there” research. As Professor Andrea Goldsmith comments,

“Every problem that we’ve addressed in traditional wireless communications over the last three or four decades is really different now because it’s a different mode of communicating,” Goldsmith said. “As so, it opens up all of these new ways of thinking about the optimal way to design this type of communication system.”

The researchers have tackled design problems such as selecting chemicals that are cheap, easy to use, and easy to “erase”. They have published papers considering theoretical foundations of this “channel”, and its capacity (such as [1] and [2]). (No, I have not worked through these papers in detail-they are quite a bit beyond my own expertise.)

Impressive work.

So, what might this be useful for?

They suggest that it might be used for situations where electromagnetic communication is impaired, or even as a backup when power is out. It might also be an unobtrusive way to mark an object or trail for another robot to find, sort of like ants do. This might also be used in nanotechnology and inside bodies, to coordinate swarms of small robots, which are too small and low powered to use radio. Also, chemical channels should be safer for biological hosts than electromagnetic radiation, all else equal.

Very interesting.


  1. N. Farsad, Y. Murin, A. Eckford, and A. Goldsmith. On the capacity of diffusion-based molecular timing channels. In 2016 IEEE International Symposium on Information Theory (ISIT), 2016, 1023-1027.
  2. Vahid Jamali, Nariman Farsad, Robert Schober, and Andrea Goldsmith, Non-Coherent Multiple-Symbol Detection for Diffusive Molecular Communications, in Proceedings of the 3rd ACM International Conference on Nanoscale Computing and Communication. 2016, ACM: New York, NY, USA. p. 1-7.

Interesting Articles from June CACM

Catching up on the June Communications of the ACM, it was a toss up for the most interesting article.

“Neuromorphic Computing Gets Ready for the (Really) Big Time” by Don Monroe

“Neuromorphic” computing–a science fiction favorite for so many reasons–hacking brains! programming minds! the Borg!–has been languishing in the shadow of Moore’s Law: silicon based hardware has had such a great run that carbon based computation has not had much traction.

Don Monroe reports that this comparative disadvantage is fading as the screaming Moore’s curve levels off.

Crucially, researchers are not trying to recreate the brain precisely (though, of course, neuroscience is working hard to understand brains), but to use the physical and functional principles to construct workable “chips” that do something interesting.

It is now possible to create chips with neurons at the rough scale of a mouse’s brain, though much smaller (insect scale) chips can do very interesting things.

Two huge mountains have to be climbed (and the Sherpas are already in the foothills, of course.)

Obviously, one wants these systems to learn.  Ironically, many of the cleverest silicon algorithms are simulating neuro learning behavior. So we would expect a neuromorphic system that can learn to be really spiffy – small, cheap, etc.

Second, we need to be able to program the darn things, one way or another.  And it’s likely to be “another”, since Von Neumann style logic doesn’t make much sense.  Current systems are doing some things adapted from the past, that look a bit like modular programming.  The fact that it works at all is amazing to me, but clearly there is lots of blue sea here to explore.

Time for a Change by

3D printing, hell. Give me 4D printing.

Since 3D printing is now in the hands of children and grandmothers (and me), it stands to reason that it no longer is the “next thing”.  Let’s add a dimension.

Neil Savage reports on “4D Printing”

So, combining “smart materials” with 3D printing, we design objects that are printed in one piece (out of multiple materials) and then unfolds or moves or whatever.

Another science fiction favorite:  a “seed” that grows into a chair when you get it home, or whatever!

Again, lots of mountains to climb, including:

Multimatierial printing isn’t trivial: the additive processes are pretty specific to each material. Just fer instance:  try printing copper on a plastic base:  the hot copper vaporizes the plastic.  Etc.

Behavior modelling.  Not only does the CAD system have to know about the geometry and materials (we know how to do this), it has to have a really good model of the dynamics and interaction of the materials.  The last bit is, ahem, “not solved”.

(I didn’t realize that Autocad had a Bio/Nano/Programmable Matter Group.  Cool.)

Noth Makes Great Comment on Quantum Computing

The current New Yorker (March 10, 2014) has a great cartoon about quantum computing by Paul Noth.  The caption reads: “Well, your quantum computer is broken in every way possible simultaneously”.

image001
“Well ,your quantum computer is broken in every way possible simultaneously” by Paul Noth (The New Yorker, March 10, 2014).

(For sale from the company.)

This is an important update to Bob’s Laws of Computing.

#1.  If you have enough computers, some of them are broken.

#2.  If you double the number of processors, you halve the number of everything else.

#3. The Ox is slow…but the Earth is patient.

Now I have to add a 4th law:

#4.  A quantum computer is both working and broken, and you can tell which is which.

Fungal Networks

I’m long known to be an enthusiast for non-human species connecting to species appropriate computer interfaces.

Spreading the intellectual net wider, there is significant research in biological computing, connecting insects, mammals or nerve cells to computer chips, creating hybrid computational systems.

One of the more offbeat efforts is out of the Unconventional Computing Centre at the University of West England:  computing with slime moulds (Physarum).

Slime moulds are quite interesting beasts. They are robust colonies of amoeba-like microorganisms, that move and grow in pursuit of food.

Prof. Andrew Adamatzky is famous for investigating the use of these organisms to solve computation problems, such as the design of traffic networks.  (See Physarum Machines: Computers from Slime Mold by Andrew Adamatzky (World Scientific, 2010) and videos).

Basically, mould food (oat flakes) is used to set the problem, the mould is left in the dark to grow.  Then the result is imaged to record the solution.  The process isn’t blazing fast, but it is quite cheap and energy efficient.

This group is exploring commercial applications, perhaps in wearable computers.

I don’t know if these efforts are especially practical technology.

But slime moulds are very interesting entities. They are “social amoeba”; existing only in communicating, collaborating groups.

There is evidence that some social amoeba practice “agriculture”, dispersing, protecting, and harvesting edible bacteria.

Cool!

(More cool bioscience readings coming soon here.)