Quantum Computing and Decentralized Supercomputers

Companies like IBM have heralded the coming of Quantum Computing as the coming of the vehicle through which we’ll be able to solve, previously unsolvable problems. IBM uses the example of transcribing and beginning to understand, a full caffeine molecule as this process is apparently all but impossible with the current computers that exist.

To truly understand Quantum Computing in an impartial way, we need to understand where it comes from. To do this in the most complete way would require a thorough knowledge of the field of Quantum Physics. If you’re interested in doing so, by all means, go ahead and do so, but if like me, you don’t quite have the time at this point, take advantage of this piece as well as the resources below, to get started.

Most important is not to answer what all is included in the field of Quantum Physics but how it is related to the ideas that make up Quantum Computing. This will be, in effect, a continuance of our previous piece on Quantum Computing and AI.

What are the Connections?

The key differentiator that’s often suggested with Quantum Computing is that it will double the storage capacities of our computers as well as their processing powers. Judging by this, experts have also put forth the idea that we will logically discover new ways to solve problems that we haven’t yet solved, which leads back to our previously mentioned example of the caffeine molecule.

Going back to this idea of doubling as well, it comes from the law of Physics that says that subatomic particles can be in more than one place or exist in more than one state, at the same time. Judging by the science behind this concept, it is actually reasonable to say that this doubled existence equals double the power in terms of solving problems in a quicker and more efficient way than any of our computers can do today.

If all of this is true, then what exactly is stopping us from having Quantum computers today?

Roadblocks between us and Quantum Supremacy

According to companies like Microsoft, there are mathematical as well as other physical constraints between us and the period in time in which we will be able to use Quantum Computers to work faster and more efficiently than we can now.

In other words, this will be Quantum Supremacy. According to certain researchers, however, this term is somewhat of a misnomer.

The currently, widely accepted state of Quantum Computing is that our infrastructure building skills haven’t quite reached the point where they can make Quantum Computers a worthwhile investment. To explain further, we have at least minimum viable products of Quantum Computers even now. Perhaps the most well-known of them is Google’s Bristlecone, a live and active Quantum computing chip. Despite this level of success, the foundational architecture that would need to be built around Quantum computers today for them to function properly is quite the undertaking.

According to MIT and Google, the biggest hurdle in the field today is using the Quantum Computing Chips and the surrounding architecture that we do have to try to create qubits, which are the particles or pieces of data inside computers that can exist in more than one place at once. One of the most shocking revelations that exists on this subject is that superconducting circuits are required to make qubits. These special circuits have to constantly sit at temperatures which are colder than outer space. Not only has Google reportedly done this at a small, internal scale, they’ve also done it because tiny changes in sound and temperature can cause qubits to lose their double state. During this process which is called “decohering,” as qubits come out of their double existence, any calculations that they are involved in start to get progressively more filled with errors.

Therefore, if qubits and the Quantum Computers that contain them don’t stay in a controlled environment, they can fail quite easily.

Preventing Decohering

Physicists have already been working on solutions to what they are now calling, “Quantum Decoherence.”

A 2013 report stated that Israeli physicists had been observing protons as they run into atoms in order to try to better understand what happens on a scientific level during the decohering of qubits. In doing this, the team’s first goal was originally to use this greater understanding to design better atomic clocks and then move on to attempting to design Quantum Computers with this new knowledge. Their primary finding, as of 2013, was that Quantum Decoherence could be controlled by knowing and using what is called an atom’s spin state. A spin state is basically the quality at which the subatomic particles in an atom deflect off of other particles in the same way that magnets deflect off of each other.

This year, this team’s research was challenged by Morton Tavel, a physics professor at Vassar College, in a piece by Scientific American magazine that included several other scientific professionals. Tavel seemed to argue that the idea of subatomic spin is no longer truly valid and it might be better to simply focus on how subatomic particles deflect off of each other. Kurt T. Bachmann, another physics professor that was involved in the Scientific American piece, suggested that we still do not quite understand what this quality of spin is and therefore, the jury is still out on how to define it. If Bachmann’s claim is taken in connection with the Israeli work on “Quantum Decoherence,” then it just might seem that we’re back to square one, since the quality that might be controlled might be invalid.

This isn’t in fact, entirely true, since the principal conclusion of the Israel experiment was that decoherence could, “perhaps,” be controlled with the knowledge of and the ability to “take advantage of” an atom’s spin state. At this point, the question remains as to what “taking advantage of an atom’s spin state” actually means in this context. In the end, this question doesn’t seem to have been answered yet but per current reports, teams are still working on research related to the same subject. For now, we’ll leave the further discussion of this to part two of this series in order to move forwards with the relationship between Quantum Computing and projects that aim to be Decentralized Supercomputers.

The Blockchain and Supercomputers

If you’ve started with Cryptocurrencies at all, then you know a bit about the Blockchain. If you don’t know anything about it yet, then just first think about how companies work now. There’s always someone in a position that controls most or all of the firm’s data. This means the products, the transactions and even anything that the company’s workers do, in addition to sensitive customer data. With scandals like the Facebook-Cambridge Analytica scandal, as well as the Equifax Data Breach, customers have started to realize just how shaky a promise it is when they sign over their personal data to a company and the company promises to keep their data safe.

With the Blockchain, no one person has this sort of power and everyone controls his or her personal data individually. Since all data that is placed on the Blockchain is encrypted to the point of being unreadable, only people that transact with each other in some way can read each other’s data. Even in this case, it is only the data that’s necessary to prove that transactions could and did go through as well as that the users truly own the funds involved.

As we spoke about in our previous piece on the Golem Network, there are Blockchain projects that are not only touching the Artificial Intelligence industry and the Cryptocurrency industry, they are also reaching for a level of computing that we have never seen before. With Golem as well as with another one of our covered projects, Effect AI, the end game is decentralized supercomputing. Let’s go back to the Blockchain for a moment. As you may know, that’s the prime example today of a decentralized database. A decentralized supercomputer may be compared to this in that it is a network of computers with no one person exerting a majority level of control over any other people. Through connecting the processing powers of everyone involved, such networks work at essentially making today’s computers obsolete.

Golem is trying to cater to almost any type of business that might need to do a project that is beyond the capability of someone’s home computer. Therefore, it’s not just for AI teams. Effect AI is an example of a more narrowly focused type of project that has the same basic product. As we’ve discussed in our previous piece, Effect AI is for AI teams in connection with people who can possibly provide data to AI teams. As of now, they haven’t tried to reach out to any other target markets. Both projects together serve to illustrate the importance of network effects related to decentralized networks. If they don’t attract enough of the users that they are trying to attract on what we can think of as both the seller side and the buyer side, then both projects will fail. In connection with this, if both networks don’t continue to offer features and applications that their users need, then their retention numbers will most likely drop beyond repair.

What we’re getting at here is that in this space, the importance of network effects is never overstated. One Forbes contributor says that to truly succeed with network effects, one needs to first understand exactly what they are as well as exactly how to use them to your advantage. According to the same contributor, this means knowing that they are the phenomenon that happens when the value of your product or of your overall business increases with the more users that it gets. She suggests that to capitalize on network effects, one needs to design one’s business from the outset in a way that it consistently offers a differentiated and unique line of products to all of its users. Perhaps, most importantly, she seems to claim that what is most important for a business is to use a different marketing, sales, customer service, and overall user experience strategy for each of its groups of customers.

With Effect AI, this could come down to two different strategies as they are only trying to work with the average person for data gathering while trying to work with AI teams to help them build their projects. With the Golem Network, the entire process could become a whole lot more complicated in that their network claims to support any project that needs a Supercomputer.

The Coming of a New Age

Everything might sound well and good, but these networks haven’t currently caught on at scale and it’s not unreasonable to wonder why.

The easiest answer is that we’ve been used to a certain kind of computing for so long that it will just take some time for a new kind to catch on.

Another possibly more likely possibility is that, like the temperature requirements that we talked about previously, the Quantum Computing space isn’t ready to be widely adopted.

The state of experiments on the subject seems to support this and when delving in further, other pieces on the subject support the idea that understanding the structure of qubits is the major hurdle that researchers need to overcome, before Quantum Computing can truly move forward.