“It became clear that we will not be able to tackle livestock by 2021, probably across the region,” he says. “And I think it’s important, especially if you’re trying to build confidence, to find ways to go where we can go back to normal. We should not simply knock on the door of unrealistic goals, such as access to public safety. I still firmly believe that what I predicted in October, to get back to normal in the summer, will come true. ”
At the beginning of March, he filled the entire shop – he decided to give as much money as he could. “I wanted to step back and let other adopters and professionals do their job,” he says. “I don’t want to disturb this place.”
They are still monitoring and storing, conducting research and analysis — on a variety of occasions, vaccinating, and the fourth wave. “When I see anything that is bothering me or worrying that I think people are not talking, I will write,” he says. But in the meantime they are focusing on other projects, such as “Shares of YOLO, ”The stock analytics platform. His main role in the epidemic was to become a member of the World Health Organization’s expert advisory team in investigating covid-19 deaths, which he shared professionally with outsiders.
“I learned a lot last year,” says Gu. It was very eye-opening. ”
Lesson # 1: Focus on the main points
“Based on scientific knowledge, my colors have shown the importance of simplicity, which is often unimportant,” says Gu. His method of predicting the dead was simple not only in its design – the SEIR component of the machine learning system – but also its connectivity, a “down-to-earth” information system. Lower means “starting with no less bones and adding more complexity as needed,” he says. “My genre only uses posterior death to predict the future. It does not use any real source.”
Gu noted that some groups have categorized different types of cases, hospital admissions, testing, travel, mask use, comorbidities, age distribution, population, pneumonia weather, pneumonia each year, population growth, air pollution, height, smoking data, your connections, number of passengers, storage, smart thermometers, Facebook posts, Google search, and much more .
“There is a belief that if you add more to the brand, or make it more sophisticated, then the brand will perform better,” he says. “But in real situations like a pandemic, when the data is very noisy, you want things to be as simple as possible.”
“I have always thought that the death of the past is what predicts the future. It’s simple: input, output. Adding other sources can make it more difficult to produce the sound signal. ”
Lesson # 2: Reduce stress
Gu feels that he has the opportunity to address the problem without a pen. “My goal was just to follow the content of the covid to learn about the covid,” he says. “This is one of the most rewarding experiences a foreigner can have.”
But by not being a psychiatrist, Gu also had to make sure he wasn’t making the wrong or wrong assumptions. “My responsibility is to create a type of model so that they can learn my mind,” he says.
“When we discover new things that are not in line with our beliefs, we sometimes simply ignore the new or ignore it, and this can lead to problems along the way,” he says. “It really hurt me, and I know a lot of other people have done that.”
“Therefore, recognizing the potential we have and recognizing it, and being able to change what we have – to change our beliefs as new as they are against them – is very important, especially in the running as we have seen with covid disease.”
Lesson # 3: Try falsehoods
“What I’ve seen over the past few months is that anyone can say or distort the data to fit what they want to believe,” says Gu. This highlights the need for more than just making assumptions.
“For me, that is the basis of my thoughts and my sayings. I have an idea, and if that idea is true, then this is what we predict will happen in the future,” he says. “And if the assumptions are just wrong, then we have to admit that what we make is not true and change accordingly. If you do not make comparisons, then there is no way to determine if you are telling the truth or the truth. ”
Lesson # 4: Learn from mistakes
“Not all the ideas I made were right,” says Gu. In May 2020, an estimated 180,000 deaths in the US by August. “This is a lot higher than we saw,” he recalls. His tested opinion proved it was wrong— “and this forced me to change my mind.”
At the time, Gu was using a full 1% mortality rate as unchanged in the SEIR simulator. In the summer it reduced the risk of death by about 0.4% (then up to about 0.7%), her mind returned to the real race.
Lesson # 5: Involve opponents
“Not everyone can agree with my opinion, and I will accept it,” says Gu, who uses Twitter to post his comments and review. “I try to be as open-minded as I can be, to defend my position, and to argue with people. It forces you to think about what you think and why you think it is right. ”
“It comes back to confirm,” he says. “If I can’t properly defend my opinion, then I’m right, and I have to say this? It helps me to understand, in conversation with other people, how to think about the situation. it really helped me to change my race. ”
Lesson # 6: Try not to believe
“Now I have serious doubts about science – and it’s not bad,” says Gu. “I think it’s always important to ask for results, but in the right way. It’s a good line. Because a lot of people are attracted to science, and I’m not the one to do that.”
“But I think it’s also important that you don’t just trust the scientists quietly,” he continues. Scientists are not perfect. “It is appropriate, he says, if something seems wrong, to ask questions and get explanations. “It is important to have a different perspective. If there is anything we learned last year, then no one is always right. ”
“I can’t speak for all scientists, but my job is to reduce all the noise and get to the truth,” he says. “I’m not saying I have a problem in the last year. I’ve been wrong many times. But I think we can all learn to use science as a way to find the truth, not the real truth. ”