Quantum computing is the future… eventually | Jason Ball | TEDxOIST
HTML-код
- Опубликовано: 18 июн 2024
- Jason Ball is a PhD student in quantum information science, a former teacher, and the father of a budding young scientist, Jason has a passion for teaching physics. Although many people are intimidated by physics, Jason believes that everyone is capable of understanding the guiding physical principles of our world, including quantum mechanics. Jason’s primary research interests at OIST lie in developing technologies to protect and transfer quantum information, and he has also taken up quantum programming as a hobby. When he’s not in the lab or the classroom, Jason can be found spending time with his family or painting miniatures. This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at www.ted.com/tedx Jason Ball is a PhD student in quantum information science, a former teacher, and the father of a budding young scientist, Jason has a passion for teaching physics. Although many people are intimidated by physics, Jason believes that everyone is capable of understanding the guiding physical principles of our world, including quantum mechanics. Jason’s primary research interests at OIST lie in developing technologies to protect and transfer quantum information, and he has also taken up quantum programming as a hobby. When he’s not in the lab or the classroom, Jason can be found spending time with his family or painting miniatures. This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at www.ted.com/tedx
I'm in a superstate. I understand it and at the same time I don't :D
Until someone observe you and give you a quiz lol...
Hahaha
you're is quantum state
The 70 years from 1946 to 2017 cannot be compared to the timespan from 2017 to 2021. Now things proceed faster, and scientists can exchange information using internet, and there are big companies that can make big investments in QC (any computing corporation in 1946?). Also, you can perform the same calculation multiple times so you reduce the risk of error.
You'd have to perform it more and more as the error rate increases. Thus the speed won't improve.
Yeah simply running it over and over again doesn't scale well. You eventually need to be able to correct errors in your qubits as they appear.
You have to realise that there has already been decades of effort to build quantum computers. Given the rate of progress, there likely needs to be many more decades of effort before quantum computers can do general computation (not very specialised tasks) faster than normal computers.
But..... we are now dealing with quantum madness....
@@regarrzo This is already what happens in a "standard" digital communications system. Any kind of discretized information gets eventually transmitted in the form of a real, continuous signal - a voltage, a current, etc., and all of these are subject to thermal noise and interference with other signals, problems which in principle affect any system and transmission - making any transmitted symbol potentially prone to errors such that a wrong bit is received and protocols like Ethernet detect that error which make the whole packet get dropped - and, if under TCP, *rentransmitted* => performing "calculations" all over again.
Moreover, at the very beginning, it was very hard to have a successful digital transmission system since additional problems like intersymbol interference were not dealt with yet (taking into account premature signals filters in receivers); it was kinda the same situation as we have now for quantum computing and quantum systems communication. It looked like the future, and yet, in some decades this new science is working very well, thanks to heuristics that govern physical constraints (preferring energy consumption over bit error rate or preferring lower bit error rate over energy etc) and, most importantly, research on error correction codes, which drastically improve the performance of a system with minor overhead (at least, for today's new technologies for which bandwidth isn't much of a big problem anymore), getting closer and closer to the Shannon's limit (which - in any case - exists for any transmission system).
I believe similar advancements will take place in a few years for quantum computing, once researchers go beyond the 5G research (honestly, given the current internet status, all those new algorithms and data structures used for 5G's new paradigms are kinda a waste of resources, but oh well). Actually, there are already some protocols for quantum communication which effectively teleport (or even dense) information with error problems being very attenuated; they worked efficiently for small distances (see for example dense coding) and the only problem is, scaling the whole thing for big distances, since the classical forward error correction codes can't be used in quantum communication due to the inability of creating redundant qubits (non-cloning theorem in quantum mechanics), however some other error correcting schemes are currently being worked on. In fact, all that error stuff is true in the sense that sharing an entangled state between a TX and a RX and keeping the superposition across big distances leads to a higher error probability than current classical communication network, buuuut, if we consider a non-distributed computing system, then distances are negligible even at the quantum level and the error rate in a classical physical medium (buses of a processor) is very low, when communicating information from a device A to a device B within the same machine. As such, the actual problem is in a possible "quantum internet" (which was already modeled), and quantum processors are already a thing. Why don't we have them yet? Because we would see their full potential in distributed computation systems, and because not being able to communicate information efficiently to the rest of the world would be pretty useless in 2019. If I recall correctly, the maximum distance obtained is something like 1400km for quantum teleportation, which is not bad (even though this requires very specialized media and optical fibers are a must, since this was obtained through photons' entanglement states) - honestly, I don't think this distance - and in general errors - problem won't be dealt with within a decade.
Universal quantum computers that highly overcome classical computers in many practical areas are certainly not around the corner, but NISQ (Noisy Intermediate Size Quantum) computers are already here. They mainly serve to emulate quantum mechanical systems, "eventually" this can speed-up very significantly improvement on quantum technology shortening the path to quantum advantage in other fields.
Are u a electrical engineer @alberto gonzale, tellez?
Hopefully, you're wrong
Very clear!!
This guy is so great at explaining!
No he isn't! He glossed over every important fact that is useful in the understanding of quantum
2:14 Both of these things are true at the same time, and you need to use quantum mechanics to explain it. Ain't that convenient.
Both are in a superposition state where both are true in some proportions, time will be the filter that will give us the answer to if a quantum computer is feasible or not.
@@happydawg2663 I've paused the video just to look for a comment like this :) I was sure someone else thought it too
My question it's possible we find time particles and dark energy and dark matter with quantum computing
Where my IonQ investors at!!!
So much talk for so little content!
My brain 🧠 is a quantum computer
My brain is an analogue computer....
But it has a 100% rate...
It has the weakest computational power but the lowest error rate
I agree.
The human brain is a binary system no quantum processes
Got a question. Maybe Jason, my nephew, will see it and clarify. Why doesn't gravity slow and eventually stop an electron from spinning?
it's G = m1 * m2/(r^2)*k so electron's mass is so low that gravity force is also low
It's a bizarre idea, for sure. The power of digital computers arguably derives from their binary nature; bits are either 1 or 0 (on or off) and telling the difference between the two states is a trivial matter. We could make 'trinary' computers or any number of possible states, or even make fully analogue computers, but we soon learned that sticking with two states and keeping any noise to an absolute minimum is the best way to build things if you want to really ramp up the clock speed.
Quantum computers, on the other hand, seem to thrive on ambiguity of sorts (superposition) but that necessarily means noise. I can't see how we're ever going to separate the two. A perfect qbit would be one that you never look at, which is completely useless, and QEC seems to fly in the face of the whole idea of quantum computing.
"A perfect qbit would be one that you never look at" Actually, we always want to look at (measure) a qubit, bc that's how we can determine the result of the computation. The superposition aspect is used for the actual computing, but at the end of the day, each qubit will end up as either a 0-state or a 1-state. We read the qubits' final configuration of 0-states and 1-states to get the result.
In 1969 everyone though that before 2000 we will live in other planets, but we are very far from that. Quantum seems more like a dream than a fact, controlling energy at such deep details is kind of science fiction
Occams razor?
indeed cya in 50 years
Quantum Computers seem more like an ASIC. I think they won't overtake regular computers but will compliment them. I mean, all the data from the Quantum ASIC is presented using regular computers so what's with the unnecessary duality? Likewise, why design a quantum chip to work like an x86 processor when they're better off doing the intense quantum calculation set?
yes but can you make a computers that occupies 12 dimensions ,Yes"
QUMPUTER - From now on, let's call them QUMPUTERS. Much better than the tongue twister 'Quantum Computers'!
They are even faster than distributed computing and network computing.
Depends on the application. In some cases they would be slower than classical computers. They work differently. They're not just faster cpus.
Nice flex, bro!
This is 2050 and we still don't have the Quantum Computer!
Eventually..
@Rotten Brainz More empirical, it's a spooky science. Lol
its basically statistics and probability with theory on how it works.
Quantum is eternal
It's both here and not here... at the same time
I wanna know what the system would look like, what the interface would look like.
Due to the unpredictable nature of technological innovation and the nearly infinite variation associated with technological advancement, your question is simply not answerable. Any attempt at answering (it would be pure fantasy based on nothing more than guesses in the dark.)
99 percent is good enough for me xD and if you ask several times it will become even higher right.
Sound like it will create a quantum instruction set without the need for any binary conversion.
How about the material to build quatum computer was found in East Turkistan?
Hello from E.T. -- 💙💜East Turkistan💙💜
You don't mine qubits.
@@Ken.- I think they're referring to the physical components of the computer.
@@Joe-yr1em what material only exists in East Turkistan that can't be found anywhere else? Unobtainium?
@@Ken.- I'm not sure honestly. We should ask him. theres many different quantum computer types.
@@Joe-yr1em or just Google it and find out that there's nothing uniquely special found in one little country that allows for quantum computing.
Or are you just trolling?
I don't understand ... if you have to check if the quantum computer gave you the right answer that means you already had the right answer at the time you asked the question, and if you already had the right answer that means you already calculated it elsewhere... and if you already calculated elsewhere it then why do you need a quantum computer at all??
After all, I assume the value of a quantum computer will only come for the unsolvable problems... but having to check each answer doesn't sound very efficient...
It's called _validation,_ and it's an intrinsic part of almost all research. Obviously the idea is that in the future you won't need to keep double checking whether it's working properly, because you'll know (through validation) that it works as intended.
We are doing all of this check/validation now to make sure these quantum computers can give us correct answers to problems. But, eventually, the problems we will be sending to the quantum computers will be so complicated that no classical computer (or human for that matter) has calculated an answer to. For those problems, we won't be able to simply "check" if the answer's correct; We'll have to just trust the quantum computer. But, you build that trust by validating its results as much as possible before hand with already-solved problems.
Hope that makes sense!
Prove it.
chicken and egg comes first
Use crystals to isolate them.
Sounds more human than a computer.
F
boom
Humans are determined to build a machine that will tell them the answers to their questions about the meaning of life. Humans will build a super computer that does appear to do this. and they will fall down and worship their god. (What's really going on is that we are manifesting what is in our hearts. Machine evolution is a powerful thing, but in comparison to spiritual/biological evolution it is an ephemeral, diluted imitation.)
If you understand Quantum computing your mind becomes Quantum??
Joking
They will have to find a way to stop the quantum computers from LITERALLY exploding.
Otherwise, they will mostly remain restricted to businesses until then.
???
AFAIK no quantum computer has ever exploded. What are you on about?! :-/
nagualdesign do you know the temperature that it has to be kept in...
@@rupertacuesta Yes.
One in 10 to the 20 is much greater than one in 100; 10 and 5 % respectively !
Are you kidding !?
no its not lol. A 1 in 100 chance of winning the lottery would be 1% while a 1 in 10^20 chance would be 0.000000000000000001 % chance lol. much lower.