Guest essay by Eric Worrall
According to the big computer we are doomed to suffer ever more damaging weather extremes. But researchers can’t tell us exactly why, because their black box neural net won’t explain its prediction.
Human activity influencing global rainfall, study finds
Anthropogenic warming of climate has been a factor in extreme precipitation events globally, researchers say
Charlotte Burton
Wed 7 Jul 2021 15.00 AEST…
While there are regional differences, and some places are becoming drier, Met Office data shows that overall, intense rainfall is increasing globally, meaning the rainiest days of the year are getting wetter. Changes to rainfall extremes – the number of very heavy rainfall days – are also a problem. These short, intense periods of rainfall can lead to flash flooding, with devastating impacts on infrastructure and the environment.
“We are already observing a 1.2C warming compared to pre-industrial levels,” pointed out Dr Sihan Li, a senior research associate at the University of Oxford, who was not involved in the study. She said: “If warming continues to increase, we will get more intense episodes of extreme precipitation, but also extreme drought events as well.”
Li said that while the machine-learning method used in the study was cutting edge, it currently did not allow for the attribution of individual factors that can influence precipitation extremes, such as anthropogenic aerosols, land-use change, or volcanic eruptions.
The method of machine learning used in the study learned from data alone. Madakumbura pointed out that in the future, “we can aid this learning by imposing climate physics in the algorithm, so it will not only learn whether the extreme precipitation has changed, but also the mechanisms, why it has changed”. “That’s the next step,” he said.
…
Read more: https://www.theguardian.com/environment/2021/jul/07/human-activity-influencing-global-rainfall-study-finds
The abstract of the study;
Anthropogenic influence on extreme precipitation over global land areas seen in multiple observational datasets
Gavin D. Madakumbura, Chad W. Thackeray, Jesse Norris, Naomi Goldenson & Alex Hall
The intensification of extreme precipitation under anthropogenic forcing is robustly projected by global climate models, but highly challenging to detect in the observational record. Large internal variability distorts this anthropogenic signal. Models produce diverse magnitudes of precipitation response to anthropogenic forcing, largely due to differing schemes for parameterizing subgrid-scale processes. Meanwhile, multiple global observational datasets of daily precipitation exist, developed using varying techniques and inhomogeneously sampled data in space and time. Previous attempts to detect human influence on extreme precipitation have not incorporated model uncertainty, and have been limited to specific regions and observational datasets. Using machine learning methods that can account for these uncertainties and capable of identifying the time evolution of the spatial patterns, we find a physically interpretable anthropogenic signal that is detectable in all global observational datasets. Machine learning efficiently generates multiple lines of evidence supporting detection of an anthropogenic signal in global extreme precipitation.
Read more: https://www.nature.com/articles/s41467-021-24262-x
As an IT expert who has built commercial AI systems, I find it incredible that the researchers seem so naive as to think their AI machine output has value, without corroborating evidence. They admit they are going to try to understand how their AI works – but in my opinion they have jumped the gun, making big claims on the basis of a black box result.
Consider the following;
Amazon ditched AI recruiting tool that favored men for technical jobs
Specialists had been building computer programs since 2014 to review résumés in an effort to automate the search process
…
Amazon’s machine-learning specialists uncovered a big problem: their new recruiting engine did not like women.
…
But by 2015, the company realized its new system was not rating candidates for software developer jobs and other technical posts in a gender-neutral way.
That is because Amazon’s computer models were trained to vet applicants by observing patterns in résumés submitted to the company over a 10-year period. Most came from men, a reflection of male dominance across the tech industry.
In effect, Amazon’s system taught itself that male candidates were preferable. It penalized résumés that included the word “women’s”, as in “women’s chess club captain”. And it downgraded graduates of two all-women’s colleges, according to people familiar with the matter.
Amazon edited the programs to make them neutral to these particular terms. But that was no guarantee that the machines would not devise other ways of sorting candidates that could prove discriminatory, the people said.
…
Read more: https://www.theguardian.com/technology/2018/oct/10/amazon-hiring-ai-gender-bias-recruiting-engine
In hindsight it is obvious what happened. The Amazon AI was told to try to select the most suitable candidates, and it noticed more male candidates were being accepted for technical jobs, likely because there were more male candidates applying. So it concluded men are more suitable for technical jobs.
It is important to note this male bias in technical jobs is purely a Western cultural issue. When I visited a software development shop in Taipei, there were just as many women as men developing software. The women I have met, in Western IT shops and in that IT shop in Taipei, were just as smart and technically capable as any man. Somehow we are persuading our women not to pursue technical careers.
My point is, when scientists unleash a black box AI on a set of data, they have no way of knowing whether the output of that AI is what they think it is, until they painstakingly rip the AI apart to work out exactly how it formed its conclusions.
The climate scientists think they have discovered a significant camouflaged anthropogenic influence. Or they may have discovered a large hidden bias in their data or models. To be fair they admit there might be problems with their training data, and the climate models they use to hindcast what conditions would have been without anthropogenic influence. “… In addition, the training GCMs might be undersampling the low-frequency natural variability such as Atlantic Multidecadal variability and Pacific Decadal Oscillation. …“. This admission should have been their headline.
Until they break their black box system down, work out exactly how their AI is reaching its conclusion, and present the real method for review, the method which is currently hidden inside their AI, it seems remarkably premature to go for a big announcement, just because they like the look of their result.
When I was in school so many decades ago, I was frequently required to show my work to demonstrate how I came to certain conclusions. With anything less, the answer was considered to be little more than a guess.
Perhaps these AI systems should be required to show their work as well.
Indistinguishable from magic. Did they explain about the chicken entrails?
AI & Machine Learning – so overhyped. If you read “The Man Who Solved The Market” by Gregory Zuckerberg, basically it took about 100 top mathematicians and programmers 10 years to put together a program that would make money of 51% of trades. – and they still manually intervene sometimes. Great if you are a hedge fund, and good someone put their money on the line, but colour me skeptical on machine learning for now.
Computers can spit out whatever the programmer tells it to do .
Programmers are not scientists, just machine language experts .
The complexity of the Earths weather system with past , present and future climate cycles is far beyond any computation even with a 1000 Einstein programmers .
So they do what they are paid to do , make us all feel guilty for the world warming as we come out of an ice age because they can parallel the rise in CO2 .
The western world is in a fast state of decline , not from climate change but from all the enforcers of those that think they can change climate change .
Indeed Feynman would say wrong: “The intensification of extreme precipitation under anthropogenic forcing is robustly projected by global climate models, but highly challenging to detect in the observational record. “
There is another interesting AI image recognition anecdote involving a network that classified a husky as a wolf, so they analysed the weightings to find out what went wrong. Whereas a human might ascribe certain features a husky possesses, such as the light patches around the eyes, the AI network had trained itself to ascribe a snowy background as the defining characteristic of huskies. Unknown to the researchers – or known on a subconscious level only – the majority of photos of wolves were taken against snowy backgrounds.
This shows very real limitations with AI being used to replace humans. But even if the diagnosis of rainfall is correct, it does not answer the fundamental questions – who is responsible for the temperature increase in the first place. My guess is, if they ever used AI models to try to find anthropogenic figure prints in delta T itself, by looking at historical temperature records, they wouldn’t find it. However, success with the rainfall model might just spur them to try just that.
GIGO
From the Met office link:
“On a global average, we see that there has been an increase of an extra half-day (so one extra day every two years) where the rainfall was over 20mm/0.78inches.
The change in the length of the longest set of consecutive wet days is only around 0.25 days (so on average, the longest run is 1 day longer every 4th year).
But the increase in the total annual precipitation is over 50mm/2inches since the beginning of the 20th century.”
LOL! sorry but WTF… and your AI will tell us how much of this “catastrophic” change is due to us… yeah…totally legit.
Climate change: US-Canada heatwave ‘virtually impossible’ without warming – BBC News
‘Climate change: US-Canada heatwave ‘virtually impossible’ without warming’
Wonder if the machines spotted that?
Give me control of the AI box for the first 7 iterations and I will show you the Global Warming.
(or anything else you wish from it).
Unbelievable! Train a program to take inputs and deliver outputs you have defined. Shock! It delivers what you programmed it to.
We are in an interglacial period of an Ice Age. Past interglacial periods were much warmer
than our Holocene period, our interglacial period has been cooler than others and in coolest part of Holocene period.
Our arctic ocean is not ice free and our Sahara desert is not grasslands- global water vapor is low and 1/3 of our land is deserts.
A machine doesn’t learn. It is trained to react (according to the programmer’s wishes) to data input, via a “reward” or scoring system set up by the programmer. If the latter is biased, then the output is also biased…
A neural network based AI with a good training data set can be expected to interpolate quite well. At least that’s the experience with other data sets….
What it wont do well is extrapolate outside of what its seen…like say the atmosphere warmer than its been trained with or water vapor higher that its been trained with, or pretty much any of the expected warmer world scenarios we might expect.
“we can aid this learning by imposing climate physics in the algorithm”
wow. who needs climate physics when it does so well.
“We are already observing a 1.2C warming compared to pre-industrial levels,” pointed out Dr Sihan Li, a senior research associate at the University of Oxford.
Stopped reading there for obvious reasons. Amazing how such bright people aren’t able to understand the little ice age.
This looks like a meta analysis, of somehow combining non statistically significant trials to claim results that reflect something other than random variation. Automating this process with AI only makes it more obscure.
WOW! Completely from out of left field, a computer fed highly biased data confirms the self-fulfilling prophesy of the climate alarmist. I’m shook to my rational heels. 😏
I can detect BS. What does that make me?
How in the world can this be a true statement? The grid meshes are not small enough, TMK.
Isn’t the whole point of AI that this is unknowable. It’s not just a case of “taking it apart” to see how it works. The process is chaotic and can not be picked apart is a deterministic way.
The Amazon job filter explanation is pure hypothesis. They saw a bias in results they did not find politically convenient, so they created an “explanation” of what was going wrong and hard-wired the software not to make the same mistake again !
The Met Office will similarly program the “basic physics” into the AI and it will dutifully “discover” that it’s all due to CO2. GIGO.
This is the same process they do by tuning climate model to produce no increase when GHG are held constant then claiming they can “prove” AGW by adding GHG back in. They artfully avoid recognising that that result is a direct consequence of the initial training and tweaking, it is NOT model output. It is model input !! GIGO.
Perhaps you need to consider that women actually prefer other occupations. Women are more inclinded to be interested in people and personal interaction, men are more interested in things. That can be hypothesised to be biologically hard-wired via evolution.
Maybe they are persuading women to enter careers they do not actually prefer to do because their economy is tech based. The speculative argument works both ways.
“Somehow we are persuading our women not to pursue technical careers.”
This is the second load of shit that I have dealt with this morning.
Kinda wandered into left field there.
Pretty sure that this is a logical fallacy: “I went someplace and saw something; therefore I know a fact that is generalizable to the entire world. I can now use that fact to criticize a similar situation in another environment.”
In fact there are even more logical fallacies packed into that paragraph, in nearly every sentence.
The critique of the machine learning man-made global warming nonsense “study” would be more powerful without gratuitous PC logical fallacies.
Who will be the machines teacher ?
If the usual suspects will do the job, the machines don’t learn anything. 😀