Simulating an epidemic
Вставка
- Опубліковано 26 бер 2020
- Experiments with toy SIR models
Help fund future projects: / 3blue1brown
An equally valuable form of support is to simply share some of the videos.
Special thanks to these supporters: 3b1b.co/sir-thanks
Home page: www.3blue1brown.com
Awesome fan-made interactives:
prajwalsouza.github.io/Experi...
learningsim.itch.io/pandemic-...
Simulations by Harry Stevens at the Washington Post:
www.washingtonpost.com/graphi...
Simulations by Kevin Simler at Melting Asphalt:
meltingasphalt.com/interactiv...
Excellent visualization of each country's current growth from Minutephysics and Aatish Bhatia:
• How To Tell If We're B...
Another good interactive to see what effect various parameters have on the shape of the curve:
gabgoh.github.io/COVID/index.html
If you want to hear a mathematician/epidemiologist's summary of COVID-19, I found this MSRI talk very worthwhile:
• COVID-19: The Exponent...
Marcel Salathé on Contact Tracing:
/ 1242430736944201730
Thanks to these viewers for their contributions to translations
Hebrew: Omer Tuchfeld
------------------
These animations are made using manim, a scrappy open-source python library: github.com/3b1b/manim
Code for this video:
github.com/3b1b/videos/blob/m...
Honestly, given that the code for that video is meant for demo purposes, and not meant to reflect the true data of COVID-19, if you want to apply these ideas to the current situation I'd recommend looking for more professional epidemiological modeling tools that are less centered around animations and pedagogy and more focused on accurate predictions. For example, the Institute for Disease Modeling has a lot of models free for people to look at and play with.
---------------
Opening music:
Candlepower by Chris Zabriskie is licensed under a Creative Commons Attribution license (creativecommons.org/licenses/...)
Source: chriszabriskie.com/divider/
Artist: chriszabriskie.com/
Other music by Vincent Rubinetti.
Download the music on Bandcamp:
vincerubinetti.bandcamp.com/a...
Stream the music on Spotify:
open.spotify.com/album/1dVyjw...
If you want to contribute translated subtitles or to help review those that have already been made by others and need approval, you can click the gear icon in the video and go to subtitles/cc, then "add subtitles/cc". I really appreciate those who do this, as it helps make the lessons accessible to more people.
------------------
3blue1brown is a channel about animating math, in all senses of the word animate. And you know the drill with UA-cam, if you want to stay posted on new videos, subscribe: 3b1b.co/subscribe
Various social media stuffs:
Website: www.3blue1brown.com
Twitter: / 3blue1brown
Reddit: / 3blue1brown
Instagram: / 3blue1brown_animations
Patreon: / 3blue1brown
Facebook: / 3blue1brown
3Blue1Brown: "In most towns, people don't actually spend their days drunkenly wandering around the city."
New Orleans: "Let me tell you about Mardi Gras."
😂
@@lbognini Ireland: ... saint Patrick's day
Me : "Wait they don't?"
😂😂😂
That's just daily life over there.
"All models are wrong; some are useful." Thanks for presenting this *very useful* model!
Another way I've heard it said is 'The map is not the terrain'.
Another way of saying it: "Math is a shorthand for a theory but does not prove the theory (as such)." In this case the parameter values are assumptions and the various results are useful only to the politicians, which, moreover, strip or ignore the feedback/validation.
Very much agreed, but more so for models involving humans, and less so for natural phenomena we've studied for a very, very long time (think N-body simulations). Making this assessment is important.
@@eric.is.online I like that angle much better, for two reasons. First, mistaking a representation for the thing it represents is a widespread issue across many domains, not just mathematical modeling. Second, "all models are wrong" sounds confused to me - like you should come up with a better definition of "right" and "wrong" in the contexts of models. It's silly to say "all maps are wrong but some are useful." Many maps are "right" but the threshold for a map being "right" is not as high as reflecting literally everything about the area being mapped.
Bruh, why use a semicolon if you're gonna use it wrong?
I'm an Emergency Medicine Physician with a PhD background in modelling. I am absolutely blown away by your work combining phenomenally effective graphics with practical modelling and a patient methodical, logical argument. Even Edward Tufte would be jealous a what you've done here. Truly one of the best presentations on any topic I've ever seen in my entire life. You deserve a medal and Isaac Newton's chair.
I don't know how it panned out in Redlands, CA, but the UK screwed up by politicians overruling logicians, I'm afraid. Even now, mostly young male narcissists shop unmasked, as civil liberties don't allow questioning. A solution is they must visor up - we did a unique community visor acceptability study, and everyone can tolerate these alternatives - for the mask intolerant. Boom, as young folk say. Retired old UK medic here. Keep safe :)
Damn that was the nicest compliment I’ve ever seen. From a fellow MD myself, thank you for taking your time to write this representing our profession.
@@tim40gabby25 scared people is the problem. Older people ruining it for the young by promoting this fear. I’m disgusted by what my generation have done to the future of my kids and grandchildren.
I read that as "Isaac Newton's hair"
and absolutly unrealistic: how to detect cases? where isolate them (at site)? R depends on absolute humidity. Distancing is bullshit, because respiratory desease can travel over kilometers.
Its ridiculous how accurate most of this ended up being
coronavirus 101
right ?
I think the word is frightening tbh.
@@andyjackson3663 its not scary its math but its stupid how we could not stop this
we're even seeing a second wave in places where people are beginning to ignore social distancing rules
The 100% social distancing model looks oddly crystalline and satisfying.
with the typical errors ^^
Looks exactly like a metal lattice, with slight lines of dislocation - which would represent grain boundaries in metal. If you think about how bonded atoms are basically dots that attract each other but repel each other at closer distances so can never touch, it's pretty analagous to a bunch of repelling dots confined in a space together. This is why you get the same structures and properties in both. Funny how maths do that sometimes.
My high school chemistry teacher tried to explain entropy in a similar way. A real-world example he gave was people choosing seats in a movie theater that’s initially empty and gradually fills up. When the occupancy is low, people are unlikely to sit down right next to a stranger, etc.
ua-cam.com/video/0U_FBHKYqRk/v-deo.html
I would like to see a similar video about the supply of goods relating to differing levels of governmental price controls.
17:58 literally "it's not a bug, it's a feature"
Louis Auffret
“It’s a bug, but I liked it so much that I made it into a feature.”
Creamy Pasta
Camper van owners just spread the disease to different communities, though. It'd be like the quarantine box simulation, but instead of it being a quarantine box, you just send them all to the same community box.
AI's also find bugs in their programmed worlds ua-cam.com/video/Lu56xVlZ40M/v-deo.html
Minimizes regard for pretentiousness.
21:10 this aged really well. in that this is exactly what happened, in the US and Europe and nearly everywhere that initially suppressed it
Didn't the US suffer from like, five distinct waves? We failed to learn our lesson four different times. Amazing.
edit: and as I recall, they were like, the minimum viable definition of a wave, where we let up as soon as R dipped below 1, instead of actual cases dropping below some number.
@@ferociousfeind8538 R
@@honestwong8942 well, yeah, five instances of the U.S. almost getting a handle on the pandemic, and then seeing that R
*fact
4th wave 🔥🔥🔥
“just like the simulations!” has never felt so real
absolutely unrealistic: how to detect cases? where isolate them (at site)? R depends on absolute humidity. Distancing is bullshit, because respiratory disease can travel over kilometers.
@@igorpashev so that’s why u should wear a mask
@@igorpashev Respiratory diseases are not always airborne if thats what you mean. Also, I need sources on the last claim because ive only known airborne diseases to remain for mere hours. Airborne diseases also infect efficiently through moist droplets from one person, also made more effective within 6 feet. Wear masks, distance yourselves. Thx champ.
b
"Kissing, sneezing on each other, all that good stuff."
Sneezing on each other, as you do.
"Sneeze on me like one of your French girls."
I cracked up at that line hard. Lol.
@@NoriMori1992 I was cracking up too hard
"If they remain unnoticed and spreadable in everyone before becoming lethal."
Laughs in Plague Inc.
@@x_x5009 This is the most unrealistic part of Plague, Inc. unfortunately. Diseases mutate, yes, but they mutate independently, and IRL every mutation needs to (re-)infect people with the newer version in order to have the "new" effects. Whereas in Plague, Inc. all your previous infections mutate at once, which makes the game fun and easier, but also completely wrong.
@@amunak_ It would be interesting to make a scenario, in which you cannot evolve (or devolve) symptoms manually, just increasing and decreasing mutations and wait for it.
@@morroghaiky6580 or a scenario where every time you mutate it creates a new patient zero which has to infect everyone again, immune suppression could make it easier to infect pre-infected people.
You guys are forgetting biowarefare! I know that none of us can be certain of where this virus came from, and either way, we can't trust the CCP or anything they tell us or their people. They're already telling them that the virus came from America.
So that aside, it's my personal belief that this is a virus that's been created in a lab (in 2013- theres records of it, and it won awards). The genetic technical capabilities of the higher ups is insane, if you dont believe in aliens and backwards engineering secretly happening, then this concept will be implausible to you. The details, proofs, and reasons behind this happening dont matter for now, but if you want to learn more, Edge of Wonder has tons of videos on this and related topics.
Anyways. This virus either gets out by accident, or on purpose. The CCP is now taking advantage of the situation, and wants to use it to decimate everything, for it's own advantage. Since it's been created in a lab, I dont know what its capable of. I dont know if the virus itself is nanotechnology, and if those nanoparticles are spreadable from person to person. Maybe the virus is programmed to replicate unnatural human designed particles, and then is able to essentially have an on and off switch that can be turned remotely.
Another one of my beliefs is that, due to chinese prophecy (I /think/ that it's in the book, tui bei tu, but you'll have to check what edge of wonder said in their video), basically, this virus will do what it's doing now, and then months later, just stop out of nowhere. No trace of it, and we all go back to normal. Then, 10 years from now, it will come back again, then disappear forever. The only way this can make sense is if the virus has an on off switch like what i described. Theres a lot more details to this that I would like to explain but people dont like reading long walls of text for some reason.
It's funny because the mortality rate is actually rising like crazy so yes you're right it's getting more and more lethal. It used to be about 3% and now it's already 4.5%
There is an interesting tweak that can be done:
Separate the "removed" in dead and recovered.
Every infected has a probability of becoming dead or recovered.
Recovered become Susceptible after a long time (Due to losing the immunity or the virus mutating)
I'd go even further and separate the recovered into immune and susceptible categories. Might also include non-human factors like animal reservoirs and mutations.
if you really want to be pedantic you could make the recovery/death ratio dependent on the current number of active cases (like if hospitals are overrun). Then you could put numbers to flattening the curve lowering the death toll even if the same total number of people got infected.
This shouldn't be a PSA on "Social Distancing," this should be required viewing for Pandemics 101. Great introduction on the thought process and problem solving that can go into epidemiology. As usual, very approachable, 3B1B. Thanks again.
Yes
18:00 "It's not a bug, it's a feature!"
No Clip
Lmao GG
Its a bug *AND* a feature.
Thats us the introverts
Myer Loeb it’s not a bug it’s a virus
When your model starts rebelling…
"maybe we don't value the heroes behind them the way we should"
Well said
ua-cam.com/video/0U_FBHKYqRk/v-deo.html
Capitalism doesn't value anything that can't turn a profit today or tomorrow. That is why real heroes like that don't get the attention they deserve.
It's like IT: do your job right, and no one will know you've done anything at all.
Yes, They are the unsung heroes
Right about now you can be saying "I TOLD YOU SO!!!"
Grant, thank you for this great presentation on simulating an epidemic. These visualizations are outstanding for understanding how the numbers really work.
"nope, I"m out" 11/10 would buy happily
Already feeling the "Nope" desire.
18:00 Now that’s called thinking outside the box.
I'm fresh lolwalker
those guys are us gamers having no problem sitting at our pc for weeks with literally no real life interaction
ua-cam.com/video/0U_FBHKYqRk/v-deo.html
the box can always be expanded
Thank you for putting this out. The effectiveness of how you explained this is the best I have seen, in fact, I had to share it because now is the time to understand the impact of your observations.
Brilliant job! Well done my friend. Very informative, nicely illustrated and calmly presented. Straight to the point. No loud music, No commercial. I just wish I could find more channels like this. BRAVO !!!
Countless pie people were murdered in the making of this video.
*removed
Infinte UA-cam Commentators cannot write a Shakespeare Play, only Monkeys
Pi people? Pie people.
we will never forget the fallen
haha
18:00 It's funny how a bug made the model more realistic.
It's the preppers bugging out ;)
After all human mind is not perfectly programmed codes.
it was weird to even experience that , i guess the math package has some huge bugs in it
yeah, world of warcraft all again... (en.wikipedia.org/wiki/Corrupted_Blood_incident)
"It's not a bug, it's a feature" - every programmer at some point.
"What if people avoid contact Wich others for a while, but then they kind of get tired and stop?"
We don't have to simulate the answer to that question anymore, we have the real life data right here.
Your content is amazing and sorely needed. Thank you!
Here’s another idea to simulate: homes
There’s small boxes outside of the main box, and each of those small boxes has 3 dots assigned to them.
Each dot has a ‘role’
1 dot has a ~25% chance to travel to a main area (grocery stores, restaurants etc.) on any day. Make three ‘hubs’ in the big box, with an equal chance for the ‘wonderers’ to travel to
The second dot has a ~75% chance on any day to leave their house, but stay away from the ‘hubs’ and try to avoid other dots.
The third dot has a ~10% chance to leave their house, but they stay right outside it.
Tactical Lemon Excellent concept. I might disagree with your specs. I'm prejudice towards attempting to model a family. At that, you may have a good beginning with three adults, perhaps a married couple with an aging parent. Now we need to add two or three children, each with various levels of activities of interaction with the adults and some potential for going off short distances with the intent to inter act with other children in somewhat similar situations and possibly those children's parents or guardians. I think I 'be lost focus on modeling a single family and have started a neighborhood. Oh well, in a more comprehensive and complex model, we'd want a central city and suburbs. NYC and SF are exceptional. Urban geographers may have something to add...
I was thinking that social distancing would probably be better modeled using many many small networks (families and tight social circles) as well.
His "communities" simulation does seem to result in the most interesting and maybe useful results. Beyond the "test and isolate" result, which has been infectious disease dogma for about 400 years in the West (probably longer in China). Of course, we have much improved abilities to test and isolate now... wish we (the US) would have used them -_-
I wonder if you could take a satellite image and set up the algorithm automatically. Identify structures, and then estimate the number of people who spend time there by the surface area it takes up. In big cities this would have to take into account the height of the buildings somehow I guess.
yes
@@m3po22 great idea but no need for satellite image, the phone network is enough, more accurate and you can track contaminated, that's exactly what they did in South Korea and China
I like how the Pis are side-eyeing each other at 10:40. Like they're saying "Ew, it's YOU."
Hey Cary, nice to see you around (though not surprising).
Love your content, hope you're doing ok!
Aye it's Cary!
Yo, what is up my dude didn't expect to see you here, it seems like you're putting a lot of time into abacaba which is cool.
Nice to see my daddy in 3b1b
I do that at the grocery store now...
I ran into this because I was looking at your Bayes' Theorem videos. However, it's now 2 years later. It's remarkable (or, as someone below put it, "ridiculous," how prescient this ended up being. It's so clear which paths we followed and how it ended up being just like you predicted.
incredible. not only super fun to watch, but also simple to understand. such a shame that so many cannot wrap their heads around such a concept when this video exists.
This is calm and reasonable and smart. A video like this is not only informative, but reassuring as well. Thanks for weaponizing your math brain to help people Grant.
And thank you for doing exactly the same thing as Grant does, but with good natured snarky commentary about the Bible and history and Christianity things. You sir, are a treasure, juat like 3 blue 1 brown.
I never expected to see you here. Love Mom
The pandemic isn't real.
@@gloekgloek3046 Damage control rabbi detected.
and absolutely unrealistic: how to detect cases? where isolate them (at site)? R depends on absolute humidity. Distancing is bullshit, because respiratory disease can travel over kilometers.
I'mma steal a comment I saw elsewhere:
Actions taken before a pandemic are always "alarmist," and actions taken afterward are always "inadequate"
So true, and it's soooo easy to criticize (and profitable if you are a journalist)
I remember reading an article covering the spanish flu, where cities handled the outbreak differently. The city with strong early measures was accused of being alarmist and such, but many more people survived.
tbh it depends. in France it was 1st inadequate then alarmist + inadequate again then nothing we'll surely be all dead by the end of the pandemia
@@nomoretalk2967 That's in part because of the socio-economic factor : A lot of people are being (rightfully or not, I won't weigh in) angry with Macron, and whatever he does, they'll keep being angry. "Gilets jaunes", pensions reforms, all those people have a deep bias against him.
Case in point, 2 weeks ago, on Thursday, he declared that schools and universities would close on Monday : He was called an alarmist. The next monday, he put confinement rules in place : He apparently did it "too late".
Better to be an alarmist than a Trump. I mean, inadequate.
Amazing work. This needs to be mandatory for everyone to learn.
This video is absolutely amazing! Thanks for your great work!
11:30 The way that the social distancing makes the dots in the top left box arrange themselves in a periodic/crystalline way, even forming grain boundaries, is so satisfying to watch!
I agree, that was deeply satisfying.
I NOTICED THAT TOO OMG the lil dots were forming crystals it was adorable
@@3blue1brown It appears that the infection rate is different within a grain vs at the boundary. I was looking at this and wondering whether any effect from this or the geometry of the box could have any physical interpretation, or whether it is an artifact of how this was simulated.
Oh, a chemist in the comment section)
Hello there))
That's known as a Wigner crystal. First observed in experiments in which electrons were poured onto the surface of liquid helium. The crystal becomes unstable at high density or as the level of fluctuations increases. Nicely explained in the Wikipedia article.
This is a really good video and I really like how you've gamed out many different scenarios and shown their outcomes.
My main criticism, and it's a pretty big one, is that you've given your preferred control mechanism, isolation of infected individuals, a huge advantage by not varying the lag between infection and isolation. I don't think isolation one day after infection is realistic for nCoV2 even in a scenario of very widespread testing. I would have liked to see how quickly the benefit of isolation falls as that lag grows to 2, 5, or 7 days.
A well-structured, polite and respectful criticism that tries to constructively bring more options to the table of an important discussion. Surely a material to be explored in a similar video. Thank you
Exactly. In our country, no protocols were made to address the pandemic until many confirmed cases were already reported and thus no efforts were made for social quarantine. It took weeks until the government responded. There's also the concern of the cases being way higher than actually reported.
This is the problem that we have. It seems that covid-19 may be infections for a number of days before becoming symptomatic in the patient. Some estimates up to 14. Even if we can have proximity testing and contact tracing it would be insufficient.
With any significant number of asymptomatic infections - whether or not they remain so - only full population testing *and* isolation of those testing positive could be effective in preventing further waves.
...And in some cases may remain asymptomatic and untested. Hence the importance of acting as though we might be infected.
THIS!
Also, I have the feeling that people are being "removed" way too quickly in his simulation, plus that the R0 for the real disease was around 3...
It's a good video about infectious diseases in general, and about the effects of different measures against the spread of it.
However, my fear is that people would underestimate the neccessity for social distancing/hygiene as they might think that isolating would have been the best anyway.
Anyway, I still think this should be shared globally (I mean the video, not the disease), so that people really understand why the should keep their distance and keep their hygiene up!
Thanks you’re passion. When i watch your videos, always i’m inspired.
I want that you aren’t disappointed about youtube’s demonetization and have more confidence about your talent.
I think until now made your videos deserve more interest and that will come true. I wish these words are good sounds to you.
Excellent animation to support your takeaways. Understanding the exponential effects of a contagion is difficult. Your work makes it much easier to understand.
Thank you and your supporters for this work.
Everyone is playing checkers while Grant is playing 23 dimensional Plague Inc
Ha
It was a good game.
1000th like
in 100D chess in minecraft using only basic redstone
Do you mean he was playing : “PARADOX-BILLIARDS-VOSTROYAN-ROULETTE-TENTH DIMENSIONAL-HYPERCUBE-CHESS-STRIPPOKER!”?
17:57 The dots are going to Greenland
Covid managed to reach Greenland (and Iceland) surprisingly early. Madagascar managed to keep it out for quite a while.
Mikołaj Kuziuk only OG’s will understand.
good joke man
I'd say those dots are going to North Korea
Plague Inc player clearly invested in cold resistance and water transmission early. Well played
17:58 "It's not a bug, it's a feature"
Storytelling done by an error in your code. Nice. 👌🏽
"The most dangerous viruses are the ones who only kill some of the host"
Plague Inc. players: **sweats nervously**
I mean this strategy is effective, make it not have deadly symptoms but incredibly infectious, and youll end up with the majority infected and able to transmit. Then you throw in all the deadly symptoms, and kill everyone
The optimal plague inc strategy for mega brutal uses an aggressive amount of symptoms with a tiny bit of lethality, and then crank it up when you make bodies stay infective and after you’re in all countries.
9:52 is the main strategy in Plague Inc.
When he talked about how diceased individuals will not stay infective, I immediately thought "true, unless I evolve necrosis for my plague".
I was also thinking about that
Exactly
You mean the bad strategy.
im curious how having multiple central locations like a school and a few grocery stores will change this assuming that every dot picks 1 of the 3 to visit also it would be interesting to see how things would change if dots try to stay in clumps similar to a family staying at home
I think it dosn't matter if one person out of the family does the shopping, as he earlier talked about people in close proximity are at risk. Eventually the whole family will get it.
Just imagine the dot being a family, if one gets sick probably everyone else will as well.
QalinaCom the families are also just a similar case to the 12 different communities. Just apply many communities of size 1-7 within each community and you’ve effectively simulated a family
Model with visiting "city centers" works that way
I built a simulation that takes into account school & work (along with household life):
n8python.github.io/epidemicSimulation/
I explain it here:
ua-cam.com/video/Yo1rL0hqHAQ/v-deo.html
An exceptional piece of work. Informative and well presented. Thank you for sharing.
Thank you for making this video. It probably has saved some lives.
Day 69:
Dots became sentient, and worked together to escape
Lutfi H nice
They are PI not dot
Day 69 is always particularly auspicious for working together towards a common goal
I think the escaped dots represent the people who saw the direction the world was going and preemptively committed suicide. I'm in college and it happened to two of my friends, one of whom I was pretty close with.
Pi-Net is coming!!
Numberphile: Who are you?
3B1B: I'm you but more polished.
Numberphile talked about the underlying mathematics and how to construct the SIR models, 3B1B talked about how to experiment with those models in order to test and predict different containment strategies.
nope
They are completely different
Different videos. You could actually follow along Numberphile's model, meanwhile this is just a showcase.
AHAAHA
this is great info right here, i love to know more about statistics and graphs, especially on important things. Your videos are great, i'm glad i have subscribed to you
I love this channel... My university frequently refers to your videos as a good learning resource and I absolutely agree! (Also, I've known this channel for years, so I'm happy to see that my university approves it as a good source :) )
2 years ago, in a completely different simulation, I had my circles escaping the box when I added a repulsive force to solve collisions! I laughed a lot here. Nevertheless, many thanks for this great video (as usual). Cheers from Italy.
All the best wishes to Italy!
Stay healthy in Italy!
Take care and stay healthy
Almost like it was designed
Where and how are these simulations done? Can you please tell?
Is is just python with some libraries or is there some other specific software for this purpose?
"even when all the parameters are the same, some runs take three times longer to reach this point than others"
This deserves more than just a side note.
With agents starting at random locations and then moving around randomly, you are likely to end up with a chaotic system.
It cannot be stressed enough how crucial it is to have multiple test runs for each set of parameters.
For some parameter sets, you might end up with almost everybody infected in one run and the infection quickly dying away in another run, just due to random fluctuations.
Virus simulations (as well as economic simulations and weather forecasts for that matter) can be highly susceptible to random noise.
@@00O3O1B Yes, a nice followup would include a qualitative analysis of the official numbers and adjustments that can be added to the series of algorithms. Even though at some point all you can do is measure the overall statistics, once the fundamental parameters of the virus are accounted for then you can isolate effects of behavioral interventions such as limiting borders, etc. I for one am interested in the cumulative effect of panic shopping and hoarding on the broader population in exacerbating issues. Those who could afford to prepare would do in excess immediately when it would have an exaggerated effect on those with lower income and it took weeks for stores to enact any restrictions to buffer the impacts.
Another major factor as this goes further will be housing instability and population movement, as that increases it will negate the effectiveness of travel restrictions.
Excellent point! This is why I started all the actions in this video once there was a sufficient number of cases. For each one I showed, I rand multiple to be sure they were qualitatively the same and tried to comment when they weren't. For proper modeling (as opposed to demonstration videos), you'd, of course, want to run many and find the right way to aggregate the results into your summaries.
@@3blue1brown So does this mean we really had to destroy the American economy to achieve this? Once the tests were available (today being maybe a few days to a week where proper testing has been a bit more consistent) could we have lifted the ban like, this week (03/28), now that we know we can test and isolate effectively? It really looked like once you have a grasp on testing and isolation, you could still live your life as long as you were being super hygienic. But can we count on EVERYONE being hygienic? Did you model only partial hygiene?
@@SeraphX2 I'm not sure the US is able to test and trace it's way out of this yet, but you're right that effective testing and isolation can halt the spread and prevent second waves (South Korea and Taiwan being excellent examples of this) with much less disruption. The disruption (and economic damage) will end when the disease does, and the fastest way to do that is to catch up with the disease as soon as possible. Shutdowns and distancing are important in getting to that point.
@@SeraphX2 the model would perhaps have to take into account hygiene/isolation compliance based on infection prevalence (fear) inorder to measure the effectiveness of public policy.
Beautiful visualization . Thank you!
Awesome video!
I’d like to comment on the central hubs and the asymptomatic people.
- Central hubs: I see that there are fundamentally different types of hubs (let me call it social gathering hot spots): markets (with limited intimate contact between participants); sport events (stadiums, where after a game supporters will hug each other and go out to drink some biers in large groups); and pubs, bars,discotheques (where people get physically very close to a large number of other people). I’d believe that the R on every of these types of hot spots is very different, so to determine what kind of event should be allowed at a given phase of a lockdown phase, it would be interesting to simulate them separately.
- Rate of infection: I wonder if asymptomatic people (some of the yellow dots in your simulation), are as infectious as those that have symptoms (and sneeze and caught). Not sure how much this would change the simulations though...
Finally, I have the feeling that mobility is very hard to model accurately. When I look at the concrete cases around me, I see that people do not move socially randomly like mostly made in simulations I’ve seen. Rather, we tend have a large number of contacts inside our immediate family, then a bit less with the colleagues in same the office, finally much less frequently with our friends/sport pals, finally with the rest of the community (in supermarkets, restaurants, bars, etc). In the of the simulations, it’s as if you create another layer of boxes within each community box. Inside those smaller boxes it is very difficult to enforce social distancing, but between them it’s easier to do it (eg close all restaurants and stop sport events, etc). I would expect that one could relax the social distancing rules inside a “micro box” (eg family, close friends, maybe office colleagues) while enforcing strict rules for other interactions (eg in those social hot spots that I mentioned before) and still get similar results that one obtain by asking every single person to observe strict social distancing to absolutely everyone (outside own’s household). (Off course, risk persons need strict social distancing from everyone, in my opinion).
The purpose of all this talk... is that it’s easier for a society to follow some sort of tolerant social rules for a long time rather than very stringent (and certainly very effective) rules. If the rules are very stringent, humans can only follow them for weeks, not months, and start being careless after some time. I live in Switzerland and I can observe that after several weeks of carefully the following social distance rules, recently almost everyone started not to follow them at all...
In any case, huge appraisal for your work !
This video was so popular they started doing this in real life
Uderated comment
go up my friend
I hope nobody tried to reanimate the control case scenario lol
UA-cam likes to recommend me news videos instead. RECOMMEND THIS GUY
Watch more of his videos, at some point it'll start recommending his videos
I was recommended this video, but then again im subscribed as well so that kinda makes sense.
Wondeful video. So lovely the way you use Pi as character to express different emotions.
I am more amazed to how you put up all this into scintillating presentation rather than the content itself. We already got tons of scientific journals and publications but good visualizers are what the world lacks now.
cant agree more
this series of animations were so stunning. Never before had I seen such a great educational content. I am a general physician and I had to go through a lot to remind my patients and families to practice social distancing. I broke my back, trying to simulate the epidemic during the recent 3-weeks . and here you did it perfectly. We owe you all!
Pandemic starts
Medics: It's hell on Earth
Mathematicians: It's free real estate
People dropping dead from the virus.
Mathematicians: They are removed
LMAO why not right?
Bartooc ua-cam.com/video/I6HM12hpblI/v-deo.html
So true! Excellent simulations! Congrats...
16:17 up to 16:30 This comparison is information I was looking for some time. Thank you 3Blue1Brown.
Everyone: It's a Pandemic.
3B1B: Well, It's also math.
surprise surprise, the best tool human invented is related to everything
*Inhale*
MATH RULES
*Exhale*
ua-cam.com/video/0U_FBHKYqRk/v-deo.html
math is art
The "second wave" scenario is why it will be very challenging to determine an end date to the social distancing, shelter in place, etc strategies, that is until a COVID-19 vaccine is available to the global population.
Well....this will change our lives for the comming year atleast. There is not going to be an end very soon.
It takes like a year to make a vaccine.
@@JohnnyProductionsOfficialTM depends how many rules you want to break!
We need to just man up and all get it, so we can get back to work and the economy can recover
@@VBYTP
Sure. The millions who will die that way are all old or weak or poor or unlucky. I'm sure you won't know any of them
WOW, i Love this one!
early on, decided to share it (but first i'll watch it to the end)
then to like, and thought that"s not enough, so also to subscribe
(first time here on the channel, by "accident", not an easy "subscriber", even less to post a comment)
then i realized, sharing (helps) makes it viral
or even (hopefully) pandemic !
;)
Thanks for the hard work
and the inspiration,
it is now clearer how i want to use my programming skills
just when i needed it!
Thanks!
Thanks for this video, it is so helpful! Already sharing!
No blue dots were harmed in the making of this film.
I'd really like to see a large version of this simulation run with parameters close to what we're experiencing in real life. It'll be like looking into the future.
For example:
- 32 "communities", 4 of which will contain central hubs and 4 of which will contain a central market
- 100% are asymptomatic for 3 days; 80% will show symptoms afterwards
- 70% will be quarantined after 1 day of showing symptoms
- The infection goes away after 14 days
- The infection radius is the same as in this video
- Recovered patients are immune to future infections
- 0.005% of people can go into total isolation ("leave the box") per day
- Travel rate between communities is normal (0.2) until a certain threshold is met (in which case, travel rates will go down by 75%)
- Social Distancing rate is low (20%) until a certain threshold is met (in which case, social distancing goes up to 75%, then 95% if things get nasty)
- Infection rate is high (0.2) until a certain threshold is met (in which case, infection rates are cut in half)
- For the first time only, if the number of active cases falls below a certain threshold, travel rates, social distancing, and infection rates go back to normal
- Certain Threshold: 0.5% of the population is infected
NOTE: The virus's actual asymptomatic period can go up to 14 days, and according to zemerick13, the infection can last a lot longer than 14 days.
That .5% threshold doesn't sound good. There would be 3 000 infected people in my city, for instance.
@@supernovaw39 if 3000 were infected that would mean only 12 dead, 11 of which were about to die shortly anyway
His source code is in the description, if anyone wants to try
3Blue1Brown is far too smart to have not run that simulation...so I worry that its ("its" being Craftie!'s simulation idea, which I think is brilliant) lack of inclusion might be a sign that the results aren't....uh...valuable....at the moment.....no now that I think about it 3Blue1Brown is likely far too ethical to run that simulation and not include it in the video regardless of what it might indicate.
They should!
- Would be interesting to model three stages of contagious 1 - asymptomatic (normal mobility) 2 - symptomatic but home (greater isolation) 3 - hospitilization (isolated)
That is a great video! More people need to see this! Great job!
This is the most elegant mathematical model for 2020. Absolutely beautiful work.
Thanks.. hopefully this would inspire a few more young pups to be interested in maths.
A "full city" model would be interesting... households with a few people each, workplaces, schools, markets, hospitals, retirement homes with younger workers, quarantine zones, X% asymptomatic cases... toilet paper hoarders...
Imagine all the world's intelligence agency doesn't have a package like that in their tool suite. This means governments are being purposefully dumb to become Thanos.
simulation like that would be very time and computationally costly.
@@SouthernHerdsman I'm certain some intelligence agencies do have that capability and model it thoroughly before releasing these pathogens against opposing nations. What do you think we're experiencing right now? A mistake, or a cold, calculated attack? There will be be a long train of conflict following this. Whether it was an intentional attack, an accidental release, or a natural occurrence, this will lead to war.
@@slartybarfastb3648 Cheer up
But for it to be even more accurate, surfaces touched and air that was coughed in would have to be added.
9:15 Anyone who has played Pandemic 2 knows this. If you give your victims internal hemorrhaging immediately, you'll lose pretty quickly. If you infect the entire world with no symptoms at all, then instantly mutate organ failure, you'll win (assuming you started in Madagascar which, as we all know, is the safest place on Earth). Thank god diseases don't suddenly mutate kill-switches like that in real life.
Well, diseases *can* mutate and suddenly become more deadly, but that just makes a new strand that doesen't affect the already infected people.
Thankfully this is not exactly the case in real life. While the coronavirus already has a few slightly different strains, viruses aren't hiveminds that can mutate all at once.
you still lose because of Greenland
How about make a virus with a long incubation and non-symptomatic spread time, with a 20x flu chance of death at the end?
Greenland is the safest not Madagascar
Excellent description of 3 variables
This is really awesome. Great work and keep it up :-)
Outbreak: *starts*
Dots: Ight imma head out
Some people have a temptation to travel on airplanes when there an epidemic at 17:50
Just look at how damn fast it spread to the four corners of this earth.
Someone: *coughs* Guess I'll explore the world!
@@kaboomwinn4026 #epidemic
If you want more information on COVID-19 from someone who _is_ an expert, I found this MSRI talk to be quite good: ua-cam.com/video/MZ957qhzcjI/v-deo.html
While here we looked at what you might call an "agent-based" SIR model, if you want to see what it looks like as a set of differential equations, Ben Sparks just did a lovely video on the topic over at Numberphile: ua-cam.com/video/k6nLfCbAzgo/v-deo.html
There are a few reasons I like the agent-based model here, for demo purposes at least. It's a bit easier to understand for those who are not comfortable yet with ODEs, for one. It also conveys how things are not deterministic; no real-world curve will look as smooth as the differential equations. It also makes it way easier to ask questions and bake other assumptions into the model. Introducing things like travel or community centers into the differential equations would get very hairy very quickly. For those who want to go much more deeply into this, the Institute for Disease Modeling has a lot of models free for people to look at and play with.
Cool!
Cool!
LOVE THIS CHANNEL THANK YOU GRANT
Isn't quarantine the exact same thing as lowering the time it takes to become Removed?
my man pulled out big stats and big epidemiology brain for this
A remake of this including re-infections would be super interesting
This is the best of instruction: well thought through and incredibly well presented in layman's terms. It is a service to all. Thanks very much.
18:00 When the model is so realistic, it includes existing real life behavior that's clearly outside the initial assumptions that were strictly programmed in.
Until now, I admired 3B1B for his amazing illustration. Now I also admire him for his social contribution.
Great Tuto ,the info on this video are so Precious, thanks for sharing 👍
Wow fantastic video, it seems your bang on with the simulations.. I wish I had watched this last year would of chilled me out abit more.
As a high school math teacher, I'm making this video a recommended watch by all my student (especially those in Algebra 2, who were just learning about exponential growth functions. THANK YOU!
This video answers literally all the math related questions I've been having about this whole thing.
The Covid-19 is a HOAX. Corona-Virus has never been isolated and checked with the 3 Postulates of Robert Koch. Pasteur has treated the whole lot/world ! ALL HAS BEEN PROVEN!!!! Social distance is DEADLY for society...
One of the three best videos on Covid19 on UA-cam (Edit:The other's are Kurzgesagt's and your other video)
Thank you 3B1B
1:49 anyone gonna talk about how cute that pi are?
Man 3Blue1Brown coming in clutch during these quarantine hours
Wasn't expecting to feel such a wide range of emotions watching some pi symbols and dots shuffling about the screen. And genuinely funny moments too (the dots leaving the Bay Area lols) Great video!
Thank you for your continued excellent and most helpful work.
you gave me an idea to make a vid. thank you so much!
1:32 I like how those Pi's become sad whenever they get infected.
Mustafa Malik
Most of us feel miserable when ill, don’t we?
☹️
Sir, honestly: the simulations and analyses were so well done and have so much common sense that I tilt my hat to your work in gratitude! I am definitely becoming a patron!
is that hat a fedora by any chance?
Thank you. Really nice. I like a simulation like this the way I used to like novels which do more than just entertain, i.e. gets things gist about right.
Very informative and comparable simulations. I look forward to the vdo that you will explain the equations behind these simulations.
I love how the 100% social distancing case created a crystal
The "escaped dots" around17:50 are the people leaving big cities.
They are the people leaving where ever they are currently. You shouldn't move anywhere for any reason.
I'd say they're more like preppers retreating to their bunkers.
@@AngleSideSideThm Preppers heading for bunkers are less risky dots than dots isolated at home. They won't infect anyone or get infected, because they are REALLY staying put. They aren't even going to grocery stores.
Going to Florida
@@Ricardo-C Your fascism is showing.
I’m sharing this video! It’s great information!
Amazing Work! Congratulations!
18:04 "It's not a bug, it's a feature" you nailed it
21:43 : "I'm writing this during a pandemic." That just hit hard.
Why, you didn't know already? We're on legal lockdown here in the UK
@@StevenMartinGuitar You think you're special? All of Europe is in lockdown. -That's how you sound, by the way.
@@Thisisahandle701 He just live in UK, so he say that he is on legal lockdown, on UK. What's the problem ? You think you're special ? In India population is on lockdown, and so for the majority of people in the world. As far as I know, there is 3 billions of persons in lockdown( ~ 700 millions people in Europe)
@@miguelgazquez5717 not the point lmao
did anyone else pick a dot and then see if they would survive
They are too small for my phone plus it’s hard to track them
No but I might try it now.
You should only do so if you're older than 75 otherwise this is no different to the flu
@@markwilson2421 hey bro, you dropped your shitty opinion in this unrelated comment’s replies
@Tahmid alpha12 you right, thx
Phenomenal video. Very well done and very informative.