I don't think I've ever watched a SH video and not come away with a substantially better understanding of whatever she's talking about. She's an absolute gem of a communicator.
Being too old to have learned this in college, I have always been amazed at the lack of clear explanation of this aspect of quantum computing. This video is the best of the dozens of videos I have watched and articles I have read. Bravo!
@@obsidianjane4413exactly what I was thinking. Don't get me wrong I'm not some anti-science nut, and I'd rather funds get misused by academics than, say, by the industrial-military, but it sours the publics' perception of academia when all they hear about is a bunch of egg head's wild goose chases.
No one is too old to learn anything. A lot of people don't want to waste their time learning something that they're not interested in.Getting a motivation to learn something is a challenge. But the reality is learning is a never-ending process. You can only stop learning when you're dead.
I 100% agree about the lack of a clear explanation, and I have always believed the old maxim, if you can't explain it clearly, you don't really understand it. The only reservation I have about not continuing to pursue it is that it may be one of those things in science that are not linear and one day there is a massive advance in the field which reaps great benefits.. However, at the moment, it doesn't seem to be clear where it's heading at all. On the matter of the super position collapsing once it's observed to a binary value, that's the first thing I wondered when I was told of the concept of the Qbit; that at some point the data will end up being represented as a function of boolean algebra.
I really appreciate how deliberate you were with your delivery in this video - especially in the intro section, your efforts to be clear and leave time for points to sink really really came across; you aren’t here to excite us or swindle us, you have no interest in sweeping information under the rug or speeding past less desirable concepts. Thanks for all you do!
An excellent choice of subject. This can't get enough attention. I have been suspecting for a long time that the expectations for quantum computing were overblown but the last 2 years or so I was getting some doubts. This was a nice refresher.
@@Vile_Entity_3545 Sentient AI doesn't have to happen. You just have to convince people it's sentient, which from what I've seen is pretty easy to do.
i looked at a lot of people explaing quantum.... you are the only one who explains it that a grunt like me can understand it, and your also honnest, you dont sell yourself or your principels, you cut trough there crap as a hot knife trough butter. i respect the human you are and how you do this work, we need way more people like you!
@@DanielMasmanian I didn't even notice it after reading this comment, watching the clip, reading the comment 3 more times, and finally understanding that finanance isn't supposed to have that many an's in it.
Watching this channel is always both intellectually challenging and somehow comforting at the same time. BTW at 8.44 "Finance " seems to have acquired an extra cubit or is somehow showing some entanglement if I am not mistaken.😉
Such a great coverage of the so-what of quantum computing for technical people who aren't physicists, and appropriately humble as to the certainty of the perspective. Instead of being mired down in the technical details of how quantum computing works, this video gives enough of an idea, but focuses on professed applications to be skeptical of, applications to be hopeful of, as well as how the exploitation of quantum computing will likely manifest itself initially.
I have been waiting for such a video! Thanks Sabine! OT: It will be alway a mystery to me how you manage to 1) do research at university 2) publish high quality videos twice a week 3) spend time with your family!
Very interesting and well presented. Having been a "Travelling Salesman", and managed inventory in warehouses, I think I'm qualified to provide a real-world response to the idea of using quantum computing to calculate the ideal solution to those problem. Yes, it can and and will do so, "...but at the end of the day..." it's most likely not going to make a world-changing improvement? Why? Because of diminishing returns, and the corollary to the idea of "diminishing returns", which is "other factors". Diminishing returns: Having planned my daily sales routes using different software packages, I quickly came to the conclusion that it is far simpler -- and just as effective -- to avoid planning bad routes. Mathematically -- if there are N! possible routes for N stops to be made, there are usually just a few "bad" routes. For example always moving from one stop to the furthest available next stop. Don't do that! Once you have eliminated the bad routes, the rest of the routes often have very similar values, and the downside risk of picking any of them is very small. So "at the end of the day", it doesn't matter which you pick since the difference between them may only be a few minutes of travel time. And given that any particular customer visit may have a variance of 10-15 minutes, saving a minute in travel time will not make a difference in the time that you arrive back at the hotel at the end of the day. This brings me to the "other factors". In a world where optimization yields diminishing returns, other factors can quickly trump the benefits of sophisticated optimization techniques. In the example given above, a talkative customer, or maybe some unexpected news of a potentially big deal (which is why I was visiting customers in the first place!) may cause me to chuck my day's plan in order to spend more time with that customer. In the case of warehousing, you may spend countless hours calculating the ideal locations for parts based on their size, weight, value and projected activity level, only to be faced with stocking the next fad product (Look! Pet rocks! A Mood ring! Slime!!!) which will suddenly cause you to throw the whole scheme out and start again. It is impossible to predict the next fad product, in either the commercial realm, nor in industrial products. There is a world of unknowns out there waiting to jump up at some unpredictable time and ruin your ideal solution. So maybe the best solution is to not get too heavily invested in any solution, and instead focus on avoiding the stupid solutions. This could be summed up by saying "Just don't do anything stupid". There are evolutionary benefits to just not getting yourself fired/killed/eliminated, and living to fight another day.
You are assuming human deliverers. Drones are already being used to deliver things now. They don't follow the same routes and they don't chat with customers
Excellent. The same will happen in the financial realm. Any economic agent that cracks the code using quantum computing will be followed by late to the game agents that also would be able to leverage the new technology, effectively nullifying the initial advantage .
@@JC_923 you seem to be thinking on a wildly smaller scale. maybe a drone can deliver your monthly supply of trash bags and toilet paper to your house, but they are not and likely will not ever be a replacement for the massive truck distribution networks that many parts of the world rely on. this comment is about warehouse scale, not consumer scale
When I read your comment, I thought "What? A video sponsored by a LaTeX platform?" (though it make sense for the audience). This is the first time I've skipped the content and gone straight to watch the ad--out of sheer curiosity. For my needs, a local latex install "works fine" after I figured out which of the trillion pieces I needed to create the documents I want to make, and it's easy enough to use git. The Overleaf documentation has been extremely helpful getting started with latex, however. The platform's collaboration features must be extremely useful for teams preparing for publication.
@@quillaja Exactly, I've completely given up trying to get people that run different Linux distros to somehow get the same Latex installation and just switched to getting everyone to use Overleaf.
I prefer plain TeX over Latex. Been using it for more than two decades. Some wonderful programs like TeX are everlasting. Thanks to Dr. Knuth for that program and to Sabine for her informative and beautiful videos.
@@douwehuysmans5959 I used the texlive "distribution" since it's kinda sorta easy to install on debian from apt. And I put it on Win10/11 too since sometimes I'm stuck using windows for ONE application. I haven't encountered any problems with working on a document while switching between platforms (yet?) but I'm probably doing fairly simple stuff.
17:22 Minor correction, factorial is not the product of all integers these then the value, it is the product of all positive numbers less then the value (otherwise the factorial would be trivial for all positive numbers, by being the number 0).
I use a self-hosted Overleaf system on my own server (inside a virtual machine - VM). It works like a charm and with the advantages of VMs, I have backups of my entire manuscripts collection. So far, I have nothing but good things to say about it. Some of my colleagues use the paid version and the open-source that I run, but I find it very surprising how many others still exchange files over email despite me telling them about Sharelatex in the past and Overleaf more recently.
I wonder why more people don't use git, or at least github, in these scenarios. I suppose because it's one more thing to learn, and git isn't super intuitive either.
@@quillaja Unless you work in computer science I don't think many people find git all that useful. In order to get the most out of git you have to use the cli. And with all the different branches, merge conflicts, local vs remote, I can easily see people not wanting to bother.
@18:30 I think one notable example of the 'travelling salesman problem' is the routing of very large scale integrated circuits having billions of elements that need to be interconnected in complex ways while being kept as short as possible and occupying least space on the chip.
I don't know if anyone else has mentioned this but quantum computing _could_ speed up the internet a little and maybe moderately by improving packet routing. Packet routing is another, "traveling salesman," sort of problem. The shortest distance between two nodes might not be a straight line. Congestion is a factor, for instance.
Ok , that would be nifty ... Still i bet youtube won't bring back buffering while the video is paused , But now even with steam powered connection we'll be able to get stuff done !
I think we can do way more than that. I've heard we can use superposition to "teleport" data i.e transmit data node to node if you will; although, technically I shouldn't use the word "transmit" because data would be in both nodes at the same time
Maybe so, @Bob Ross. That was kinda part of my statement, wasn't it? You never know what kind of clever algorithms someone will come up with, though. Have you seen the lengths people go to to find HW and SW bugs? Would it have occurred to you to write to the memory rows on each side of a row to change the data in the target row? And that's a fairly obvious one once you hear it. (The actual implementation of Rowhammer is pretty complex, as I understand it, though.) Yes, I see what you're saying about needing to know what's going on to figure out the best route but who knows what will come up?
Thank you for being the voice of sobriety in the frat party of quantum computing. I did a little quantum computing last year to learn more about it, and you have finally revealed some of the basic truths I discovered. Of course, I was only working with logical qbits, which was problematic enough. As a seasoned computing professional, I cannot help wonder if quantum computing will scale as well as people hope, where scaling problems are rarely appreciated until people start hitting speed bumps, or drive the car off a cliff. On the other hand, I can appreciate all the fun people are having, as I too know what it's like to be young and drunk at a frat party.
Hi Dr. Hossenfelder! I am absolutely loving your videos, and I’m so glad you cut through some of the ego-driven noise of men in physics-finally feels like things make sense without trying to posit a special ingredient to humanity or consciousness. In my free-time, I’ve been trying to construct a theory surrounding an impressionistic language of neuro-transcription given specialized neuronal sensitivity and collapsed the body/mind equation into a total sensory self. I don’t expect you to simply engage with me based on a comment, especially seeing the liveliness of everyone else clearly resonating with your work as well. That being said, I am struggling with what to do as this working theory builds meaning as I interview neurologists and psychiatrists, even meshing with your beautifully concise explanation of super-determinism. I hope one day I have the time to go back to school and develop this with other working scientists. Thank you for all you share. I would never have found an access point into quantum physics without it!
Ordinarily I don't comment on the sponsor message, but Overleaf sounds really useful -- like it finally has addressed a long standing problem of complex formulae typesetting (with collaboration!). I definitely will give this one a try!
I really wish i could support your channel financially. Originally, my partner and i started watching because we loved listening to your voice/accent, but we found we're actually understanding shit and learning new stuff. We really appreciate you and what you do. You make us feel like students again - but this time we love the teacher!!
Wonderful to see such a realistic evaluation of quantum computers, thank you very much. I would also be interested in seeing how travalling salesman or vehicle routing can be formulated for quantum computers. Shorr's algorithm for factoring numbers is already pretty complicated (_and_ somewhat probabilistic), so I wonder how these optimization problems would turn out and how one would convert the result back to the non quantum world.
What happens when your competitor uses it to take you down quickly? If you deliver in these scenarios, the computer can then cause you problems. A faked breakdown at an intersection, a truck with flat tires, a railroad with a bad stretch of tracks. Just because we can do something does not mean when should. Oh I see you have the computer and he does not so you preemptively take him down. Where is humanity in this?
Sabine you slay me! 😂 Great info and wonderful humor. Thanks! BTW, I used to work for IBM and had heard that United Parcel Service was using neural nets to figure out their delivery routes. Training the neural nets with years of data of actual routes figured out by humans.
@ Sabine Thank you for answering something that was confusing me. When we measure the wave collapses. 3m12. Now I understand a bit better how we can make sense from a Quantum computer calculation.
Thank you, Sabine, for another excellent video. I have to admit that reading about QC's in the media, and some predictions, confused me. This video has clarified things for me. And your humour about physicists taking over this and that...please tell them to continue and maybe they'll eventually get to 'quantum politics'. Jeez, they certainly couldn't do any worse than the mess politicians seem to be in today, 😊
Hi Sabine, The TSP problem is NP-Complete, so there is actually not likely much boost that quantum computing will give over classical solutions. Any papers on a quantum speedup for NP-Hard problems has been over naive brute force solutions.
Hey Sabine - GREAT Channel..! A SUGGESTION: Could you perhaps do a video explaining - The Archimedes Experiment - and it's attempt to measure the Casimir Effect and it's relation to vacuum energy. Thank You -Cheers
Thanks, it is an excellent video. As in many new technologies, the press tends to over-promise. Could you comment on the applications of quantum computers in machine learning? I understand that methods based on Grover's search algorithm which has been shown to solve unstructured search problems with a quadratic speedup compared to classical algorithms. Another exciting topic is Quantum annealing for optimisation is not a universal quantum computer but many engineering problems are optimisation problems.
The jokes in-between make me laugh so much I have to keep rewinding bits of the video over and over to capture all the serious stuff. It’s a perfect unrivalled combination of humour, dare and science. Absolutely loving it and thank you for doing this.
Thanks for the suggestion... Actually this video was supposed to be about quantum simulations, but I got distracted along the way. I'll give it another try!
Wow , i happened to binge watch a lot of your videos . U r like my school teacher - simple, elegant, humorous, straight to the point, especially in this confusion era of knowledge
Can we maybe have a video about how a quantum computer is actually build? I always see these fancy looking images of them and hear a lot of theory about it, but what are the actual components, how do make the operations/entaglements and how do you put information in and out?
In this video, you seemed to hint that we understand more about how to predict material properties based on electron efficiency (If I didn't misunderstand, which is completely possible lol), could you do a video on what we know about predicting these properties? Thank you for all the lovely videos!
Good info, I like you mentioned some examples of what we could do with quantum computers. Most videos just explain what it is in as short of time as possible.
If today's computer engineering students find Verilog and VHDL challenging, they have Q-loads of fun waiting for them. When we first brought up next-gen iron, the very first thing we'd use it for is to speed up simulations of it's next revision. Noticed no mention of D-Wave and its quantum annealing. Fair enough.
The topic of quantum computing came up recently at work and I'm very tempted to share this video with my colleagues, as I truly believe this is the best educational video I've seen so far on quantum computing. But the frequency in which underwear was mentioned gave me pause.
I can do facility location selection well without needing quantum computers. Same with traveling salesmen problems, GIS software have heuristic algos that do this quite well.
All I want to know is - when can I get a spaceship armed with a Warp Drive, Temporal Shielding and QUANTUM TORPEDOES with the Temporal Upgrade package that allows the torpedo phase out of ordinary space, penetrate enemy shields and explode INSIDE of the enemy Wessel?
Ive never seen anyone speak as comfortably about quantum physics, mechanics and computing as you are in your content. It's impressive and very inspiring given how lost the quantum topics make me feel. All the best to you! P. S. I wish your husband a swift recovery from this sick burn.
I really L💞VE your channel you make the nerdy information funny😂 and fun🎉 your channel is my most 👍favorite👍 Channel on RUclips now keep up the good work I love it.🌎✌
I should start every day with your videos. Great information, well explained and, of course, very funny quips about life. It gets me in a good mood every time
Not only is there a lot of room at the bottom, there is also a lot of fascinating material properties to be exploited in the vast area between the physical qubit and current nanoscale CMOS
When I was introduced to TeX and LaTeX (mid to late 1980’s) my professor told me it was pronounced “LAH-tekh” and that’s how I’m used to hearing it pronounced. Checking the Wikipedia page, “LAY-tekh” is also listed as an acceptable pronunciation, but I think Sabine is the first person I’ve noticed pronouncing it that way. I wonder if there will be a shift to this “new” pronunciation like the shift of GIF to JIF.
JIF has never been correct! 🤡 But seriously, I do think it comes down to the fact that, in English, a hard G usually precedes a short I like "gig" and "gift"), while a soft G usually precedes a long I (like "giant" and "angina"). So the best case for a J sound is if it sounds like JAIF, and I hope we can agree that sounds weird.
@@sabinrawr I realize now I wrote GIF and JIF backwards from what I intended. When the GIF standard was first published I was still in college and I distinctly remember reading the documentation stating that it was “pronounced jif like the peanut butter” (“Choosey developers choose Jif” was a joke back then). But developers sometimes loose control of their works, including pronunciation (for example, see Linux) and I agree with you that a hard G, like the G in Graphics (since that is what the G stands for) is reasonable, and certainly has become the common usage.
@@jpe1 Okay, you're forgiven. You're one of the good eggs! Though, to this day, I have a hard time imagining what the reasoning behind the J pronunciation, assuming that the joke wasn't also the punchline.
Thank you for the amazing explanation. I love the way you broke it down into Logical Quibits and Physical Quibits, A lot of quantum computer hype and explanations do a poor example of explaining how we can use the quantum properties to actually perform calculations. As well as how this compares and works in comparison to traditional computers, transistors, and logic gates.
Sabine, I surrender to your incredibly dry humour particularly your unexpected references to fancy underwear inter"laced" (pardon the pun:-) ) with your brilliant observations on and crystal clear explanations of the scientific subjects and phenomena you are presenting us with ! You are a breath of fresh air in the sadly often stuffy and stifling world of scientific discovery and development. Wow , I often joke with my best mate that I have fallen in love with you and the way you stir my imagination... uniquely due to your intriguing persona and how you do what you do so well! 🥰 🏆💡🧲 Man Overboard!! 🤸♀🤸♂
This may have been the first sponsorship where my reaction was: "I've been using this for years and it's actually great!" So congratulations on being sponsored by overleaf!
Note: I heard some quantum scientists say that they are very eagerly looking for future Quantum programmers and they encourage people to take advantage of the open source projects.
@@LordZordid I work on a quantum computing project for drug discovery. Sabine is right. Even if we can prove QC advantages, there is nothing useful QC can do beyond niche applications. It's even more specialized than GPUs. And nobody knows how to scale up quantum computers.
I remember when the BBC had a TV programme in the late 90s about the world's first quantum computer and its amazing how far but also how slow the engineering for it is
Thanks for the clear and concise overview of Quantum Computing. I agree it’s probably a long time off, which is great for all those biologists and financiers who won’t have to worry about physicists coming for their specialty.
Hey Sabine - not entirely related, but I noticed that the captions during your Overleaf section gave a targeted link ("/sabine"), while on-screen you only mentioned their base URL. I want to make sure you get credit where credit is due for conversions!
Hello Sabine!👋 Love your Content and Humor!🥰 I find pleasure in expanding my knowledge and understanding of how we currently believe the existence we are experiencing, functions?😁🤔 Thanks for all your hard work in bringing us such Interesting and Diverse subject matter!👍 Perhaps your husband could benefit from a Quantum computer Pizza selection App!😁 Oh no, that's right, He is Non-linear. Nevermind.🤭 Mike in San Diego.🌞🎸🚀🖖
I was working on a school project about designing transport routes, and the hypotheses I tried to prove turned out wrong. Then I thought of forgetting optimization and instead regularly updating existing routes based on new data, but I couldn't figure out what parameters to use. Ultimately I gave up on that and chose another project. So I understand what you mean in the "Logistics" section.
To get an idea of the formal definitions and what happens to computational complexity: An ideal quantum computer would be one with an infinite number of (logical) qubits and in which each imaginable entanglement between those (logical) qubits is possible. If such a machine is possible, it essentially becomes a Non-deterministic Turing machine: en.wikipedia.org/wiki/Nondeterministic_Turing_machine The notable difference between a deterministic and a non-deterministic Turing machine is that the deterministic Turing machine can have only exactly one output of it's transition function, while a non-deterministic one can have a set as output of its transition function and can magically choose the right state to transition to from that set of potential answers (this happens at waveform collapse). Reality however, is much more constrained. It's impossible to have an infinite (but an amount that can be considered "infinite" for all practical purposes might be possible) number of (logical) qubits and geometry makes it impossible to build a chip on which all possible entanglements can be set up. Therefore you'll end up with something more constrained like what's described here: en.wikipedia.org/wiki/Quantum_Turing_machine So some of the problems in NP will get easier, but they cannot be solved in O(1), but they will be solvable in O(sqrt(b)) where b is the number of qubits required to model the input and outputs of the problem if the entire problem can be modeled in one entanglement-cycle. If you need repeated iterations over the same qubits to solve your problem, you will need to run O(sqrt(b)) quantum calculations for each iteration you'll need. On top of that, you'll also need to repeat the entire circus multiple times, because just one answer from a quantum-chip does not tell you anything. You'll need to obtain statistical distributions over the final states of all your input and output qubits if you want your answer to tell you something meaningful. So: It'll be a far-cry from the exponential time we'd need on classical computers, but it won't be the next miracle technology and it will still take time regardless. The real questions are: "Can quantum states be preserved long enough to make repeated iterations possible?" and "Can initialization be done fast enough in a consecutive manner such that it can beat the billions of calculations a classical computer can do, without thermal noise ruining your quantum states?" Oh and on code breaking: Quantum computers are more or less useless in cracking symmetric block ciphers. So your WiFi-network and many card-based payment transactions cannot be cracked any faster by using a quantum computer. But what can be cracked is most current cryptography that uses public key exchange mechanisms like SSL/TLS (HTTPS), SSH and so on.
This video comes with a quiz that will help you remember what we talked about: quizwithit.com/start_thequiz/1687349958014x203376479823740960
Thanks Sabrina 🥰
Sabine, hallo
I have not seen you're videos in 2 years much love you are so photogenic and physical in the entanglement of the day happy solstice.
I don't think I've ever watched a SH video and not come away with a substantially better understanding of whatever she's talking about. She's an absolute gem of a communicator.
Then please don’t see her video on capitalism. You’re too impressionable.
@@nothingissimplewithlloydWdym.. Sabrina doesn’t get involved in silly political issues like that…
Yeah, Pretty nice rack as well 😂
Being too old to have learned this in college, I have always been amazed at the lack of clear explanation of this aspect of quantum computing. This video is the best of the dozens of videos I have watched and articles I have read. Bravo!
Because the more obscure it is, the more grandiose they can make the promises and the larger the grant request bottom line can be.
@@obsidianjane4413exactly what I was thinking. Don't get me wrong I'm not some anti-science nut, and I'd rather funds get misused by academics than, say, by the industrial-military, but it sours the publics' perception of academia when all they hear about is a bunch of egg head's wild goose chases.
physics girl did a decent video of it
No one is too old to learn anything. A lot of people don't want to waste their time learning something that they're not interested in.Getting a motivation to learn something is a challenge. But the reality is learning is a never-ending process. You can only stop learning when you're dead.
I 100% agree about the lack of a clear explanation, and I have always believed the old maxim, if you can't explain it clearly, you don't really understand it.
The only reservation I have about not continuing to pursue it is that it may be one of those things in science that are not linear and one day there is a massive advance in the field which reaps great benefits.. However, at the moment, it doesn't seem to be clear where it's heading at all.
On the matter of the super position collapsing once it's observed to a binary value, that's the first thing I wondered when I was told of the concept of the Qbit; that at some point the data will end up being represented as a function of boolean algebra.
I really appreciate how deliberate you were with your delivery in this video - especially in the intro section, your efforts to be clear and leave time for points to sink really really came across; you aren’t here to excite us or swindle us, you have no interest in sweeping information under the rug or speeding past less desirable concepts. Thanks for all you do!
0 seconds ago
369 thieR are no reminders
⚛⚛🕉🕉🕉🕉🕉🕉🕉🕉⚛⚛
You can’t really determine that from one video
As a biochem student, this is the first ad on youtube that I did not skip.
An excellent choice of subject. This can't get enough attention. I have been suspecting for a long time that the expectations for quantum computing were overblown but the last 2 years or so I was getting some doubts. This was a nice refresher.
It is like sentient AI and fusion energy. It may never happen.
@@Vile_Entity_3545 Sentient AI doesn't have to happen. You just have to convince people it's sentient, which from what I've seen is pretty easy to do.
ciaopizzabella - To reiterate for my own edification - your doubt about your skepticism has been refreshed?
2:27
@Heywood I meant that now I'm convinced again that the expectations are overblown
i looked at a lot of people explaing quantum.... you are the only one who explains it that a grunt like me can understand it, and your also honnest, you dont sell yourself or your principels, you cut trough there crap as a hot knife trough butter. i respect the human you are and how you do this work, we need way more people like you!
If there's one thing you've taught me, its cautionary optimism! Thanks for the informative video
Or enthusiastic pessimism, whichever you prefer (or don't)! 😁
I love Overleaf! I used that for almost all of my homework while getting my math degree, it basically got me through university
Finanance (8:47) "No spell-check in LaTeX", ChatGPT said to Clippy, and they both laughed their AA's off.
Ah, sorry, none of us noticed!
@@SabineHossenfelder no, but some of *them* did
@@DanielMasmanian I didn't even notice it after reading this comment, watching the clip, reading the comment 3 more times, and finally understanding that finanance isn't supposed to have that many an's in it.
@@kindlin that's understandable, as financiers aren't famous for noticing much at all
@@DanielMasmanian I'd like to hand you the medal for best comment of the day! 😂
Cloud LaTeX sponsoring a YT video, what a time to be alive.
Watching this channel is always both intellectually challenging and somehow comforting at the same time.
BTW at 8.44 "Finance " seems to have acquired an extra cubit or is somehow showing some entanglement if I am not mistaken.😉
Not even quantum computing can add a cubit to your stature.
@@michaelwright2986 I am stealing that one - maybe a bit niche but pure gold.
Such a great coverage of the so-what of quantum computing for technical people who aren't physicists, and appropriately humble as to the certainty of the perspective. Instead of being mired down in the technical details of how quantum computing works, this video gives enough of an idea, but focuses on professed applications to be skeptical of, applications to be hopeful of, as well as how the exploitation of quantum computing will likely manifest itself initially.
Sabine is fantastic - her humor, especially!
Also kinda smart
trillions of years, like how long it takes her husband to decide on what pizza 😂 I honestly spat out my beer. It's great we can rewind
@@paulgoogol2652 lol just KINDA, lol.
@@altareggo kinda
Almost as fantastic as a pair of fancy underpants
You had a superposition of letters in the word Finanance.
I love your videos, just got turned onto this channel. Great work!
I have been waiting for such a video! Thanks Sabine!
OT:
It will be alway a mystery to me how you manage to 1) do research at university 2) publish high quality videos twice a week 3) spend time with your family!
I rely heavily on my team.
@@SabineHossenfelder impressive nevertheless! Have a nice day 🙂
@@SabineHossenfelder Admit it, you do know quantum family mechanics and won't share that. 😆
@@SabineHossenfelder Me wondering if the 'team' wrote this comment...
Interesting... women are likely to be questioned about how they balance their work and family obligations. Such remarks are rare if it's a man.
Very interesting and well presented. Having been a "Travelling Salesman", and managed inventory in warehouses, I think I'm qualified to provide a real-world response to the idea of using quantum computing to calculate the ideal solution to those problem. Yes, it can and and will do so, "...but at the end of the day..." it's most likely not going to make a world-changing improvement? Why? Because of diminishing returns, and the corollary to the idea of "diminishing returns", which is "other factors".
Diminishing returns: Having planned my daily sales routes using different software packages, I quickly came to the conclusion that it is far simpler -- and just as effective -- to avoid planning bad routes. Mathematically -- if there are N! possible routes for N stops to be made, there are usually just a few "bad" routes. For example always moving from one stop to the furthest available next stop. Don't do that! Once you have eliminated the bad routes, the rest of the routes often have very similar values, and the downside risk of picking any of them is very small. So "at the end of the day", it doesn't matter which you pick since the difference between them may only be a few minutes of travel time. And given that any particular customer visit may have a variance of 10-15 minutes, saving a minute in travel time will not make a difference in the time that you arrive back at the hotel at the end of the day.
This brings me to the "other factors". In a world where optimization yields diminishing returns, other factors can quickly trump the benefits of sophisticated optimization techniques. In the example given above, a talkative customer, or maybe some unexpected news of a potentially big deal (which is why I was visiting customers in the first place!) may cause me to chuck my day's plan in order to spend more time with that customer. In the case of warehousing, you may spend countless hours calculating the ideal locations for parts based on their size, weight, value and projected activity level, only to be faced with stocking the next fad product (Look! Pet rocks! A Mood ring! Slime!!!) which will suddenly cause you to throw the whole scheme out and start again. It is impossible to predict the next fad product, in either the commercial realm, nor in industrial products. There is a world of unknowns out there waiting to jump up at some unpredictable time and ruin your ideal solution. So maybe the best solution is to not get too heavily invested in any solution, and instead focus on avoiding the stupid solutions. This could be summed up by saying "Just don't do anything stupid". There are evolutionary benefits to just not getting yourself fired/killed/eliminated, and living to fight another day.
You are assuming human deliverers. Drones are already being used to deliver things now. They don't follow the same routes and they don't chat with customers
Excellent. The same will happen in the financial realm. Any economic agent that cracks the code using quantum computing will be followed by late to the game agents that also would be able to leverage the new technology, effectively nullifying the initial advantage .
And the traveling salesman problem already has good enough algorithms (heuristic ones) for many cases.
@@JC_923 you seem to be thinking on a wildly smaller scale. maybe a drone can deliver your monthly supply of trash bags and toilet paper to your house, but they are not and likely will not ever be a replacement for the massive truck distribution networks that many parts of the world rely on. this comment is about warehouse scale, not consumer scale
Thank you for spreading the good word of latex and overleaf! I’ve been using it for awhile and have been super happy with it!
When I read your comment, I thought "What? A video sponsored by a LaTeX platform?" (though it make sense for the audience). This is the first time I've skipped the content and gone straight to watch the ad--out of sheer curiosity.
For my needs, a local latex install "works fine" after I figured out which of the trillion pieces I needed to create the documents I want to make, and it's easy enough to use git. The Overleaf documentation has been extremely helpful getting started with latex, however. The platform's collaboration features must be extremely useful for teams preparing for publication.
@@quillaja Exactly, I've completely given up trying to get people that run different Linux distros to somehow get the same Latex installation and just switched to getting everyone to use Overleaf.
I prefer plain TeX over Latex. Been using it for more than two decades. Some wonderful programs like TeX are everlasting. Thanks to Dr. Knuth for that program and to Sabine for her informative and beautiful videos.
@@douwehuysmans5959 I used the texlive "distribution" since it's kinda sorta easy to install on debian from apt. And I put it on Win10/11 too since sometimes I'm stuck using windows for ONE application. I haven't encountered any problems with working on a document while switching between platforms (yet?) but I'm probably doing fairly simple stuff.
@@mathewpv681 I'm a troff kind of guy myself. 🙂
16:45 is fascinating to me as a first responder humanitarian assistance / disaster response logistician -thank you!
I absolutely love Sabine's Jokes and analogies.
Please don’t ever stop with the humour. 😂 Makes it so much more fun learning new stuff!
17:22 Minor correction, factorial is not the product of all integers these then the value, it is the product of all positive numbers less then the value (otherwise the factorial would be trivial for all positive numbers, by being the number 0).
17:22*
@@amannvig Thanks
Meteorologists are laughing in the face of physicists knowing they will never take over meteorology.
I use a self-hosted Overleaf system on my own server (inside a virtual machine - VM). It works like a charm and with the advantages of VMs, I have backups of my entire manuscripts collection. So far, I have nothing but good things to say about it. Some of my colleagues use the paid version and the open-source that I run, but I find it very surprising how many others still exchange files over email despite me telling them about Sharelatex in the past and Overleaf more recently.
I wonder why more people don't use git, or at least github, in these scenarios. I suppose because it's one more thing to learn, and git isn't super intuitive either.
@@quillaja Unless you work in computer science I don't think many people find git all that useful. In order to get the most out of git you have to use the cli. And with all the different branches, merge conflicts, local vs remote, I can easily see people not wanting to bother.
@18:30 I think one notable example of the 'travelling salesman problem' is the routing of very large scale integrated circuits having billions of elements that need to be interconnected in complex ways while being kept as short as possible and occupying least space on the chip.
Sabine. You are amazing and you should have your own show or podcast. The world needs to hear what’s in your brain
Like a weekly talk show with guests and panel discussions? Yes, absolutely!!
I don't know if anyone else has mentioned this but quantum computing _could_ speed up the internet a little and maybe moderately by improving packet routing. Packet routing is another, "traveling salesman," sort of problem. The shortest distance between two nodes might not be a straight line. Congestion is a factor, for instance.
Ok , that would be nifty ...
Still i bet youtube won't bring back buffering while the video is paused ,
But now even with steam powered connection we'll be able to get stuff done !
I think we can do way more than that. I've heard we can use superposition to "teleport" data i.e transmit data node to node if you will; although, technically I shouldn't use the word "transmit" because data would be in both nodes at the same time
Maybe so, @Bob Ross. That was kinda part of my statement, wasn't it?
You never know what kind of clever algorithms someone will come up with, though. Have you seen the lengths people go to to find HW and SW bugs? Would it have occurred to you to write to the memory rows on each side of a row to change the data in the target row? And that's a fairly obvious one once you hear it. (The actual implementation of Rowhammer is pretty complex, as I understand it, though.)
Yes, I see what you're saying about needing to know what's going on to figure out the best route but who knows what will come up?
Printed circuit board routing is another possible use. Multi-layer, many point-to-point and serially connected points to consider.
The traveling salesman problem, yes
The consideration on quantitative finance is actually really cool.
Sabine, has some awesome considerations
Sabine, your intellect inspires me and your optimism drives me. Thank you for what you do❤
Thanks!. You are keeping my old brain in good shape.
Thank you for being the voice of sobriety in the frat party of quantum computing. I did a little quantum computing last year to learn more about it, and you have finally revealed some of the basic truths I discovered. Of course, I was only working with logical qbits, which was problematic enough. As a seasoned computing professional, I cannot help wonder if quantum computing will scale as well as people hope, where scaling problems are rarely appreciated until people start hitting speed bumps, or drive the car off a cliff. On the other hand, I can appreciate all the fun people are having, as I too know what it's like to be young and drunk at a frat party.
Hi Dr. Hossenfelder! I am absolutely loving your videos, and I’m so glad you cut through some of the ego-driven noise of men in physics-finally feels like things make sense without trying to posit a special ingredient to humanity or consciousness. In my free-time, I’ve been trying to construct a theory surrounding an impressionistic language of neuro-transcription given specialized neuronal sensitivity and collapsed the body/mind equation into a total sensory self. I don’t expect you to simply engage with me based on a comment, especially seeing the liveliness of everyone else clearly resonating with your work as well. That being said, I am struggling with what to do as this working theory builds meaning as I interview neurologists and psychiatrists, even meshing with your beautifully concise explanation of super-determinism. I hope one day I have the time to go back to school and develop this with other working scientists. Thank you for all you share. I would never have found an access point into quantum physics without it!
Ordinarily I don't comment on the sponsor message, but Overleaf sounds really useful -- like it finally has addressed a long standing problem of complex formulae typesetting (with collaboration!). I definitely will give this one a try!
Good job to Sabine for making complex topics understandable to the average person! 📚
Overleaf is great. Especially combined with using Mathematica to write the equations and output latex with TexForm
I really wish i could support your channel financially. Originally, my partner and i started watching because we loved listening to your voice/accent, but we found we're actually understanding shit and learning new stuff. We really appreciate you and what you do. You make us feel like students again - but this time we love the teacher!!
Wonderful to see such a realistic evaluation of quantum computers, thank you very much. I would also be interested in seeing how travalling salesman or vehicle routing can be formulated for quantum computers. Shorr's algorithm for factoring numbers is already pretty complicated (_and_ somewhat probabilistic), so I wonder how these optimization problems would turn out and how one would convert the result back to the non quantum world.
I think if there was a quantum algorithm to solve the NP-complete problems, it would have gotten a lot of publicity and we would know about it.
What happens when your competitor uses it to take you down quickly? If you deliver in these scenarios, the computer can then cause you problems. A faked breakdown at an intersection, a truck with flat tires, a railroad with a bad stretch of tracks. Just because we can do something does not mean when should. Oh I see you have the computer and he does not so you preemptively take him down.
Where is humanity in this?
Heh, "Finanance" instead of "Finance" in the list at 8:40. Nice to see that Sabine (or at least her video editors) are not infallible!
Sabine you slay me! 😂
Great info and wonderful humor. Thanks!
BTW, I used to work for IBM and had heard that United Parcel Service was using neural nets to figure out their delivery routes. Training the neural nets with years of data of actual routes figured out by humans.
@ Sabine
Thank you for answering something that was confusing me. When we measure the wave collapses. 3m12.
Now I understand a bit better how we can make sense from a Quantum computer calculation.
Sabine, Thank you for this video. It gives clarity.
Thank you, Sabine, for another excellent video. I have to admit that reading about QC's in the media, and some predictions, confused me. This video has clarified things for me. And your humour about physicists taking over this and that...please tell them to continue and maybe they'll eventually get to 'quantum politics'. Jeez, they certainly couldn't do any worse than the mess politicians seem to be in today, 😊
"Do you guys just put 'quantum' into everything?!" :D - smartest line in MCU.
Hi Sabine, The TSP problem is NP-Complete, so there is actually not likely much boost that quantum computing will give over classical solutions. Any papers on a quantum speedup for NP-Hard problems has been over naive brute force solutions.
Overleaf looks amazing! 🙂 Especially in comparison to my beginnings in LaTeX in high school - in MS-DOS 😅 Printed out manual was a must 😅
Hey Sabine - GREAT Channel..! A SUGGESTION: Could you perhaps do a video explaining - The Archimedes Experiment - and it's attempt to measure the Casimir Effect and it's relation to vacuum energy. Thank You -Cheers
Glad you do this series, Usually this topic and others usually are presented with Plans for a rosy future; hype.
While discussing the TSP you showed the map of one of my favourite cities in the world, very good choice👌
Thanks, it is an excellent video. As in many new technologies, the press tends to over-promise. Could you comment on the applications of quantum computers in machine learning? I understand that methods based on Grover's search algorithm which has been shown to solve unstructured search problems with a quadratic speedup compared to classical algorithms. Another exciting topic is Quantum annealing for optimisation is not a universal quantum computer but many engineering problems are optimisation problems.
The jokes in-between make me laugh so much I have to keep rewinding bits of the video over and over to capture all the serious stuff. It’s a perfect unrivalled combination of humour, dare and science. Absolutely loving it and thank you for doing this.
Great video as always! Could you possible do a video on quantum simulation?
Thanks for the suggestion... Actually this video was supposed to be about quantum simulations, but I got distracted along the way. I'll give it another try!
Thank you for keeping it light and relatively transparent
Thanks for the lucid video on quantum computers! Your videos are informative, entertaining, and funny!
Wow , i happened to binge watch a lot of your videos . U r like my school teacher - simple, elegant, humorous, straight to the point, especially in this confusion era of knowledge
Can we maybe have a video about how a quantum computer is actually build? I always see these fancy looking images of them and hear a lot of theory about it, but what are the actual components, how do make the operations/entaglements and how do you put information in and out?
Yes, it's very mysterious. I'd also like a fairly detailed explanation of how the software/operating system is designed.
An excellent, informative video and all people find important to point out is that "Finance" is spelled "Finanance" at 8:44
In this video, you seemed to hint that we understand more about how to predict material properties based on electron efficiency (If I didn't misunderstand, which is completely possible lol), could you do a video on what we know about predicting these properties? Thank you for all the lovely videos!
I used Overleaf to write both of my thesis papers, great tool!
Quote of the day:
"If you compute with something that doesn't exist, we call that politics." -- Sabine Hossenfelder
hahah yeah she had me dead with that one
@@two2truths That means you don't exist, and youtube computed with you, so youtube is political!!!!
Best line!!!😂😂😂🥰
Good info, I like you mentioned some examples of what we could do with quantum computers. Most videos just explain what it is in as short of time as possible.
If today's computer engineering students find Verilog and VHDL challenging, they have Q-loads of fun waiting for them. When we first brought up next-gen iron, the very first thing we'd use it for is to speed up simulations of it's next revision. Noticed no mention of D-Wave and its quantum annealing. Fair enough.
Thank you Sabine. It’s great to find another person using RUclips sharing actual information. I love your videos.
Always looking forward to an episode!
Is that from Getty stock quotes?
The topic of quantum computing came up recently at work and I'm very tempted to share this video with my colleagues, as I truly believe this is the best educational video I've seen so far on quantum computing. But the frequency in which underwear was mentioned gave me pause.
L😂L!!! Just warn them…
I love the idea of shouting factorials. I shall definitely do that from now on.
Seen on. a T-shirt in the 80s:
Mathematicians know 2 + 2 ≠ 4!
I wonder if the Aliens 👽 know much more about primes?😂
I can do facility location selection well without needing quantum computers. Same with traveling salesmen problems, GIS software have heuristic algos that do this quite well.
All I want to know is - when can I get a spaceship armed with a Warp Drive, Temporal Shielding and QUANTUM TORPEDOES with the Temporal Upgrade package that allows the torpedo phase out of ordinary space, penetrate enemy shields and explode INSIDE of the enemy Wessel?
Ever heard Space Peace by Droids? Go and listen to it!
'Computing with things that don't exist', thank you for the new definition!
Ive never seen anyone speak as comfortably about quantum physics, mechanics and computing as you are in your content. It's impressive and very inspiring given how lost the quantum topics make me feel. All the best to you! P. S. I wish your husband a swift recovery from this sick burn.
I really L💞VE your channel you make the nerdy information funny😂 and fun🎉 your channel is my most 👍favorite👍 Channel on RUclips now keep up the good work I love it.🌎✌
I should start every day with your videos. Great information, well explained and, of course, very funny quips about life. It gets me in a good mood every time
Have expectations, and no expectations at the same time.
Progress is slow, until everything changes.
Not only is there a lot of room at the bottom, there is also a lot of fascinating material properties to be exploited in the vast area between the physical qubit and current nanoscale CMOS
When I was introduced to TeX and LaTeX (mid to late 1980’s) my professor told me it was pronounced “LAH-tekh” and that’s how I’m used to hearing it pronounced. Checking the Wikipedia page, “LAY-tekh” is also listed as an acceptable pronunciation, but I think Sabine is the first person I’ve noticed pronouncing it that way. I wonder if there will be a shift to this “new” pronunciation like the shift of GIF to JIF.
JIF has never been correct! 🤡
But seriously, I do think it comes down to the fact that, in English, a hard G usually precedes a short I like "gig" and "gift"), while a soft G usually precedes a long I (like "giant" and "angina").
So the best case for a J sound is if it sounds like JAIF, and I hope we can agree that sounds weird.
@@sabinrawr I realize now I wrote GIF and JIF backwards from what I intended. When the GIF standard was first published I was still in college and I distinctly remember reading the documentation stating that it was “pronounced jif like the peanut butter” (“Choosey developers choose Jif” was a joke back then). But developers sometimes loose control of their works, including pronunciation (for example, see Linux) and I agree with you that a hard G, like the G in Graphics (since that is what the G stands for) is reasonable, and certainly has become the common usage.
@@jpe1 Okay, you're forgiven. You're one of the good eggs!
Though, to this day, I have a hard time imagining what the reasoning behind the J pronunciation, assuming that the joke wasn't also the punchline.
Thank you for the amazing explanation. I love the way you broke it down into Logical Quibits and Physical Quibits, A lot of quantum computer hype and explanations do a poor example of explaining how we can use the quantum properties to actually perform calculations. As well as how this compares and works in comparison to traditional computers, transistors, and logic gates.
Sabine, I surrender to your incredibly dry humour particularly your unexpected references to fancy underwear inter"laced" (pardon the pun:-) ) with your brilliant observations on and crystal clear explanations of the scientific subjects and phenomena you are presenting us with ! You are a breath of fresh air in the sadly often stuffy and stifling world of scientific discovery and development. Wow , I often joke with my best mate that I have fallen in love with you and the way you stir my imagination... uniquely due to your intriguing persona and how you do what you do so well! 🥰 🏆💡🧲
Man Overboard!! 🤸♀🤸♂
The more episodes from this brilliant teacher I am watching, the more I am thrilled. THIS is world class edutainment! 🥰
Wrap me in a Photonic Prophylactic and call me Pamela.
Grovers search algorithm could cause huge speed ups if the oracle function is nice to us.
This may have been the first sponsorship where my reaction was: "I've been using this for years and it's actually great!" So congratulations on being sponsored by overleaf!
Note: I heard some quantum scientists say that they are very eagerly looking for future Quantum programmers and they encourage people to take advantage of the open source projects.
It doesn't work that way, though.
@@DJVARAO I am paraphrasing.
@@LordZordid I work on a quantum computing project for drug discovery. Sabine is right. Even if we can prove QC advantages, there is nothing useful QC can do beyond niche applications. It's even more specialized than GPUs. And nobody knows how to scale up quantum computers.
Great video, awesome to hear some other potential applications besides cracking RSA
Great informerS Sabine! I'm wondering how likely it is that the "AI" ability growth could solve these problems with quantum computers?
I love the sense of humor. Excellent presentation.
10:09 Glad to see some of the most important problems acknowledged. Hopefully quantum computers can reduce pizza ordering time
Just yesterday I watched someone to show how AI (current chat gpt) order pizza for you via bot calling on your behalf..
I remember when the BBC had a TV programme in the late 90s about the world's first quantum computer and its amazing how far but also how slow the engineering for it is
Thanks for the clear and concise overview of Quantum Computing. I agree it’s probably a long time off, which is great for all those biologists and financiers who won’t have to worry about physicists coming for their specialty.
Hey Sabine - not entirely related, but I noticed that the captions during your Overleaf section gave a targeted link ("/sabine"), while on-screen you only mentioned their base URL. I want to make sure you get credit where credit is due for conversions!
Thanks Sabine. This is the only explanation of quantum computing that ever made any sense to me.
"We call that politics" LOL!!
Thanks for revisiting the topic!
Hello Sabine!👋
Love your Content and Humor!🥰
I find pleasure in expanding my knowledge and understanding of how we currently believe the existence we are experiencing, functions?😁🤔
Thanks for all your hard work in bringing us such Interesting and Diverse subject matter!👍
Perhaps your husband could benefit from a Quantum computer Pizza selection App!😁
Oh no, that's right, He is Non-linear.
Nevermind.🤭
Mike in San Diego.🌞🎸🚀🖖
I liked your Overleaf commercial a lot. Will definitely give it a try.
"Don't believe the hype."
"Here's more hype."
I think you nailed it.
I was working on a school project about designing transport routes, and the hypotheses I tried to prove turned out wrong. Then I thought of forgetting optimization and instead regularly updating existing routes based on new data, but I couldn't figure out what parameters to use. Ultimately I gave up on that and chose another project. So I understand what you mean in the "Logistics" section.
No references?
Overleaf is the probably the best sponsor I've ever seen on RUclips.
Investing in crypto now should be in every wise individuals list, in some months time you'll be ecstatic with the decision you made today
To get an idea of the formal definitions and what happens to computational complexity:
An ideal quantum computer would be one with an infinite number of (logical) qubits and in which each imaginable entanglement between those (logical) qubits is possible.
If such a machine is possible, it essentially becomes a Non-deterministic Turing machine: en.wikipedia.org/wiki/Nondeterministic_Turing_machine
The notable difference between a deterministic and a non-deterministic Turing machine is that the deterministic Turing machine can have only exactly one output of it's transition function, while a non-deterministic one can have a set as output of its transition function and can magically choose the right state to transition to from that set of potential answers (this happens at waveform collapse).
Reality however, is much more constrained. It's impossible to have an infinite (but an amount that can be considered "infinite" for all practical purposes might be possible) number of (logical) qubits and geometry makes it impossible to build a chip on which all possible entanglements can be set up. Therefore you'll end up with something more constrained like what's described here: en.wikipedia.org/wiki/Quantum_Turing_machine
So some of the problems in NP will get easier, but they cannot be solved in O(1), but they will be solvable in O(sqrt(b)) where b is the number of qubits required to model the input and outputs of the problem if the entire problem can be modeled in one entanglement-cycle.
If you need repeated iterations over the same qubits to solve your problem, you will need to run O(sqrt(b)) quantum calculations for each iteration you'll need.
On top of that, you'll also need to repeat the entire circus multiple times, because just one answer from a quantum-chip does not tell you anything. You'll need to obtain statistical distributions over the final states of all your input and output qubits if you want your answer to tell you something meaningful.
So: It'll be a far-cry from the exponential time we'd need on classical computers, but it won't be the next miracle technology and it will still take time regardless. The real questions are: "Can quantum states be preserved long enough to make repeated iterations possible?" and "Can initialization be done fast enough in a consecutive manner such that it can beat the billions of calculations a classical computer can do, without thermal noise ruining your quantum states?"
Oh and on code breaking: Quantum computers are more or less useless in cracking symmetric block ciphers. So your WiFi-network and many card-based payment transactions cannot be cracked any faster by using a quantum computer. But what can be cracked is most current cryptography that uses public key exchange mechanisms like SSL/TLS (HTTPS), SSH and so on.