N
Thoughts on the retirement of the Tupolev TU-154
- Space Turkic States to launch joint satellites
- Poland Poland got approved to buy 360 Advanced Anti-Radiation GuidedMissile-Extended Range (AARGM-ER)
- Ukraine Ukrainian mechanics have mastered the repair of US M777 howitzers.
- TECHINT Germany arrests 3 suspected China spies
- Mil Intelligence Military court convicts US sailor of attempted espionage
Breakthrough sciences and technologies
- Thread starter Combat-Master
- Start date
Zafer
Experienced member
- Messages
- 4,483
- Reactions
- Nation of residence
-
- Nation of origin
-
I was thinking everybody knew how to make one.
A Danish University has published how to build your own drone... lol, sorry it's in danish, but google translate isn't shabby
And the shoppinglist to build one
- Messages
- 9,318
- Reactions
- Nation of residence
-
- Nation of origin
-
Sad to see her go.
I always was fond of empennage-trijet design in general....powerful but keeps the wing nice and clean.
- Messages
- 9,318
- Reactions
- Nation of residence
-
- Nation of origin
-
- Messages
- 9,318
- Reactions
- Nation of residence
-
- Nation of origin
-
This might be above most readers heads here....but this is extremely significant (having worked with eigenvector math myself w.r.t vibration analysis).
If the eigenvectors present can be determined given just eigenvalue (Scaling factors)...it suggests there is a likely inbuilt "theory of everything" after all (or one can certainly be developed substantially with time)....much like the value of Pi (and many other fundamental constants) are fixed in our universe given the manifestation of underlying geometry and physics in it.
It now basically becomes a race for more data in this field to populate correlation and credibility given immense simplification present (say with computer models) when you just handle by eigenvalues.
www.quantamagazine.org
If the eigenvectors present can be determined given just eigenvalue (Scaling factors)...it suggests there is a likely inbuilt "theory of everything" after all (or one can certainly be developed substantially with time)....much like the value of Pi (and many other fundamental constants) are fixed in our universe given the manifestation of underlying geometry and physics in it.
It now basically becomes a race for more data in this field to populate correlation and credibility given immense simplification present (say with computer models) when you just handle by eigenvalues.
Neutrinos Lead to Unexpected Discovery in Basic Math | Quanta Magazine
Three physicists stumbled across an unexpected relationship between some of the most ubiquitous objects in math.
www.quantamagazine.org
Intelligent cameras enhance human perception
Intelligent cameras are the next milestone in image and video processing A team of researchers at the Chair of Multimedia Communications and Signal Processing at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) has developed an intelligent camera that achieves not only high spatial and temporal but also spectral resolution. The camera has a wide range of applications that can improve environmental protection and resource conservation measures as well as autonomous driving or modern agriculture. The findings of the research have been publishedas an open access publication.
'Research up to now has mainly focused on increasing spatial and temporal resolution, which means the number of megapixels or images per second,' explains lecturer Dr. Jürgen Seiler. 'Spectral resolution -- the wavelength and thus the perception of colours -- has largely been adjusted to match human sight during the development of cameras, which merely corresponds to measuring the colours red, green, and blue. However, much more information is hidden in the light spectrum that can be used for a wide range of tasks. For example, we know that some animals use additional light spectra for hunting and searching for food.'
Three resolutions in one camera
Seiler, who is an electrical engineer, has therefore developed a high-resolution multi-spectral camera that enhances human perception with his team at the Chair of Multimedia Communications and Signal Processing (LMS) led by Prof. Dr. Kaup at FAU. It combines all three resolutions -- spatial, temporal and spectral -- in a cost-efficient solution. 'Up to now, there were only extremely expensive and complex methods for measuring the ultraviolet or infrared ranges of light or individual spectral bands for special industrial applications,' says Seiler. 'We looked for a cost-efficient model and we were able to develop a very cost-effective multi-spectral camera.'
The researchers connected several inexpensive standard cameras with various spectral filters to form a multi-spectral camera array. 'We then calculated an image in order to combine the various spectral information from each sensor,' explains Nils Genser, research associate at LMS. 'This new concept enables us to precisely determine the materials of each object captured using just one single image.'
At the same time, the new camera is greatly superior to existing systems in terms of its spatial, temporal and spectral resolution. As the surroundings are recorded by several 'eyes' as is the case with human sight, the system also provides a precise indication of depth. This means that the system not only precisely determines the colour and certain material properties of objects it captures, but also the distance between them and the camera.
Ideal for autonomous driving and environmental technology
Autonomous driving is a potential application for these new intelligent cameras. 'A whole range of solutions to various problems has now opened up thanks to our new technology,' says Seiler. 'In the infrared range, for example, we can differentiate between real people and signposts using the thermal signature. For night driving, we can detect animals crossing the road with sufficient warning.'
The high-resolution multi-spectral cameras could also be used for protecting the environment and conserving resources. 'Several plastics differ significantly from each other in various ranges of the spectrum, which is something the new intelligent camera can reliably detect,' Genser emphasises. 'Large amounts of plastics are simply burned instead of separated for recycling as they have a similar appearance. We can now separate them reliably.'
Intelligent cameras enhance human perception
A team of researchers has developed an intelligent camera that achieves not only high spatial and temporal but also spectral resolution. The camera has a wide range of applications that can improve environmental protection and resource conservation measures as well as autonomous driving or...
www.sciencedaily.com
IEEE Xplore Full-Text PDF:
- Messages
- 9,318
- Reactions
- Nation of residence
-
- Nation of origin
-
Hah, he is doing a countdown format....this is now the 4th worst ever I suppose:
Wonder who the top 3 worst ones will be.
Gary
Experienced member
- Messages
- 7,770
- Reactions
- Nation of residence
-
- Nation of origin
-
Little-known Korean CPU threatens to make Nvidia, Intel and AMD obsolete in HPC market
By Mayank Sharma 17 hours agoCould herald the next generation of supercomputers that are faster yet cheaper to run
(Image credit: Shutterstock / Timofeev Vladimir)
South Korea’s Electronics and Telecommunications Institute (ETRI) has team up with Arm to design a CPU that’s tailored for supercomputing applications.
The news is particularly significant in light of the fact that all of South Korea’s High Performance Computing (HPC) is powered by Intel processors, according to reports.
It is this stranglehold of US companies that fueled the ETRI research.
(Image credit: The Next Platform)
Reports say the researchers were tasked to design a CPU that was more than twice as fast as the accelerators used in the current generation of supercomputers while consuming less than half the power.
The reports quote Youngsu Kwon from the AI Processor Research Department at ETRI as saying that the researchers tackled their brief by focussing on single chip performance for low power chips and systems: “From there you can integrate more chips, increasing the performance and reducing power consumed. Also, the integration of CPUs and accelerators into a single chip will allow more bandwidth, which can remove the data bandwidth bottleneck.”
hey achieved this by combining the Arm Zeus high-performance chips with ETRI’s scalable AI/HPC cores together with multiple DDR5 high-bandwidth memory (HBM) interfaces. The result of their efforts is the K-AB21, which packs 16 teraflops per CPU for a combined output of 1600 teraflops per rack.
The group is still fine-tuning certain elements of the chip, but expects it to be available by the end of 2021.
Little-known Korean CPU threatens to make Nvidia, Intel and AMD obsolete in HPC market
Could herald the next generation of supercomputers that are faster yet cheaper to run
E
Multiple errors
There are multiple errors in your post. The paper by the four authors was still a pleasant read.
Error 1:
Update on December 4, 2019: In the weeks following the publication of this article, the researchers became aware of over three dozen places where the identity had appeared in the literature since 1966. They have now rewritten their original paper to include the history of the identity, along with all seven known proofs. As Tao describes in a blog post, they also “speculate on some possible reasons why this identity only achieved a modest amount of recognition and dissemination prior to the November 2019 Quanta article.”
Source 1
Error 2:
For large unstructured matrices, it does not seem at present that the identity (2) provides a competitive algorithm to compute eigenvectors. Indeed, to use this identity to compute all the eigenvector component magnitudes |vi,j |, one would need to compute all n − 1 eigenvalues of each of the n minors M1, . . . , Mn, which would be a computationally intensive task in general; and furthermore, an additional method would then be needed to also calculate the signs or phases of these components
Source 2
In summary, as interesting as the paper was, due to limitations of current search engine technologies and some related issues, the authors could not initially discover earlier known instances of the same algorithm or variations thereof.
Secondly, the algorithm does not simplify calculations for most matrices that may be of interest or concern. It may only be of passing interest to all but the most devoted enthusiasts in this particular subdiscipline of the already specialized discipline of Computational Linear Algebra (or some variants of it, by which it may be called in various locations).
A pleasant read, nevertheless.
There are multiple errors in your post. The paper by the four authors was still a pleasant read.
Error 1:
Update on December 4, 2019: In the weeks following the publication of this article, the researchers became aware of over three dozen places where the identity had appeared in the literature since 1966. They have now rewritten their original paper to include the history of the identity, along with all seven known proofs. As Tao describes in a blog post, they also “speculate on some possible reasons why this identity only achieved a modest amount of recognition and dissemination prior to the November 2019 Quanta article.”
Source 1
Error 2:
For large unstructured matrices, it does not seem at present that the identity (2) provides a competitive algorithm to compute eigenvectors. Indeed, to use this identity to compute all the eigenvector component magnitudes |vi,j |, one would need to compute all n − 1 eigenvalues of each of the n minors M1, . . . , Mn, which would be a computationally intensive task in general; and furthermore, an additional method would then be needed to also calculate the signs or phases of these components
Source 2
In summary, as interesting as the paper was, due to limitations of current search engine technologies and some related issues, the authors could not initially discover earlier known instances of the same algorithm or variations thereof.
Secondly, the algorithm does not simplify calculations for most matrices that may be of interest or concern. It may only be of passing interest to all but the most devoted enthusiasts in this particular subdiscipline of the already specialized discipline of Computational Linear Algebra (or some variants of it, by which it may be called in various locations).
A pleasant read, nevertheless.
This might be above most readers heads here....but this is extremely significant (having worked with eigenvector math myself w.r.t vibration analysis).
If the eigenvectors present can be determined given just eigenvalue (Scaling factors)...it suggests there is a likely inbuilt "theory of everything" after all (or one can certainly be developed substantially with time)....much like the value of Pi (and many other fundamental constants) are fixed in our universe given the manifestation of underlying geometry and physics in it.
It now basically becomes a race for more data in this field to populate correlation and credibility given immense simplification present (say with computer models) when you just handle by eigenvalues.
Neutrinos Lead to Unexpected Discovery in Basic Math | Quanta Magazine
Three physicists stumbled across an unexpected relationship between some of the most ubiquitous objects in math.
- Messages
- 9,318
- Reactions
- Nation of residence
-
- Nation of origin
-
Multiple errors
There are multiple errors in your post. The paper by the four authors was still a pleasant read.
Error 1:
Update on December 4, 2019: In the weeks following the publication of this article, the researchers became aware of over three dozen places where the identity had appeared in the literature since 1966. They have now rewritten their original paper to include the history of the identity, along with all seven known proofs. As Tao describes in a blog post, they also “speculate on some possible reasons why this identity only achieved a modest amount of recognition and dissemination prior to the November 2019 Quanta article.”
Source 1
Error 2:
For large unstructured matrices, it does not seem at present that the identity (2) provides a competitive algorithm to compute eigenvectors. Indeed, to use this identity to compute all the eigenvector component magnitudes |vi,j |, one would need to compute all n − 1 eigenvalues of each of the n minors M1, . . . , Mn, which would be a computationally intensive task in general; and furthermore, an additional method would then be needed to also calculate the signs or phases of these components
Source 2
In summary, as interesting as the paper was, due to limitations of current search engine technologies and some related issues, the authors could not initially discover earlier known instances of the same algorithm or variations thereof.
Secondly, the algorithm does not simplify calculations for most matrices that may be of interest or concern. It may only be of passing interest to all but the most devoted enthusiasts in this particular subdiscipline of the already specialized discipline of Computational Linear Algebra (or some variants of it, by which it may be called in various locations).
A pleasant read, nevertheless.
Yeah I investigated further a few days after I posted it here and came across the same stuff. I forgot to update here
....it explains why this (from 2019) didn't blow up a lot more in 2020.Thanks for giving the summary.
- Messages
- 5,157
- Solutions
- 2
- Reactions
- Nation of residence
-
- Nation of origin
-
I also didn't find out if their technique involves or handles repeating eigenvalues, or is it applicable assuming that eigenvalues are unique.Yeah I investigated further a few days after I posted it here and came across the same stuff. I forgot to update here
Thanks for giving the summary.
At least in vibration problems i deal with, i have problems regarding to the calculation of eigenvalue or repeating eigenvalues due to the axial symmetry since they become greater and appear denser (per frequency step) as frequency increases,meanwhile calculation of eigenvectors are easier due to the matrix type and condition.
E
Yeah I investigated further a few days after I posted it here and came across the same stuff. I forgot to update here
Thanks for giving the summary.
No problems.
The theorem and its generalizations have long been known and attracted minor attention because of its trivial nature.
There are lots of consumers of mathematics and then there are subject matter experts.
Mere consumers, like the three physicists in question, were wrongly led to believe this was a breakthrough. Subject matter experts would have a vastly different take on the topic for obvious reasons.
As a general rule and a first estimate, it would be prudent to say subject matter experts in Mathematics who are the at the forefront often lead mere consumers by 100 years, more or less.
The rediscovered theorem is even older since it can be derived from Cramer's rule, which is more than 2 centuries old. A rediscovered theorem such as this should not have created such big headlines IMO.
E
Repeated eigenvalues lead to a degenerate case for this theorem. Both sides of the identity equal zero, then.
A further generalization, also found in that paper accounts for repeated eigenvalues. Other degenerate cases can also be effectively handled as outlined in that paper.
If you can calculate eigenvectors, then calculating eigenvalues should be a trivial exercise even in the absence of a computer.
Difficulties may arise if you have to calculate eigenvectors for given eigenvalues even with the availability of a computer. More so for larger systems.
- Messages
- 8,145
- Reactions
- Nation of residence
-
- Nation of origin
-
Sounds like everyone learned to calculate it the difficult way. But now it can be done much faster and easier.This might be above most readers heads here....but this is extremely significant (having worked with eigenvector math myself w.r.t vibration analysis).
If the eigenvectors present can be determined given just eigenvalue (Scaling factors)...it suggests there is a likely inbuilt "theory of everything" after all (or one can certainly be developed substantially with time)....much like the value of Pi (and many other fundamental constants) are fixed in our universe given the manifestation of underlying geometry and physics in it.
It now basically becomes a race for more data in this field to populate correlation and credibility given immense simplification present (say with computer models) when you just handle by eigenvalues.
Neutrinos Lead to Unexpected Discovery in Basic Math | Quanta Magazine
Three physicists stumbled across an unexpected relationship between some of the most ubiquitous objects in math.
I'd compare it with ppl solving Pythagoras triangle using cos-relations.
Maybe I oversimplified it a bit, but goes to show only ppl who knows about it can tell how significant discovery that is
Trial and Error, and fail, fail, fail, fail....
Last edited:
- Messages
- 9,318
- Reactions
- Nation of residence
-
- Nation of origin
-
A real pity:
Though I can see why they wrote it off looking at pictures of it now.
Though I can see why they wrote it off looking at pictures of it now.
81 Pakistanis among world’s top 2% scientists
Eighty one (81) Pakistani academics have figured in the list of the world’s top two per cent scientists in a global list issued by the United States’ Stanford University. The global list, compiled by Prof John Ioannidis and his team, carries names of 159,683 people from all scientific disciplines.
All these authors have been selected on the basis of an international evaluation of their research papers and classified based on career-long citation impact until the end of 2019 as well as for their citation impact in a single year (2019).
Eighty one (81) Pakistani scientists have figured in the list for their career-long citation impact while 243 Pakistani scientists have been included for citation impact in a single year. A report on this list has been recently published in PLOS Biology.
As many as 11 teachers of Islamabad’s Quaid-i-Azam University (QAU) have figured among the top two per cent researchers. Five scientists from the University of Haripur (UoH), including its vice chancellor, are also included in the list.
The QAU academics whose names have appeared in the list include Prof Bilal Haider Abbasi of Biotechnology, Zabta Khan Shinwari and Mushtaq Ahmad of Plant Sciences, Amir Ali Shah of Microbiology and Riffle Nasim Malik of Environment Sciences.
The other names include Rashid Khan of Biochemistry, Masood Khan of Mathematics, Afzal Shah and Aamer Saeed of Chemistry and Abdul Haq of Statistics.
The UoH academics who figure in the list include UoH Vice Chancellor Prof Dr Anwar-ul-Hassan Gilani, Dr Khalid Zaman, Dr Hashim, Dr Shah Fahad and Dr Mohammed Farooq.
UoH Vice Chancellor Prof Gilani is the only serving VC in Pakistan who has appeared in this list and who has been conferred three civil awards including Hilal-i-Imtiaz.
Three professors of the University of Punjab also figure in the list.
These are Prof Dr Muhammad Sharif of Nuclear and Particle Physics, Prof Dr Khalid Mahmood of Economics and Management Sciences and Prof Dr Muhammad Akram of Artificial Intelligence.
Talking to APP, QAU Academic Staff Association General Secretary Prof Dr Bilal Abbasi, whose name is also included in the list, said it is an honour for the entire country that a number of Pakistani scientists working in different universities have been named in top two percent scientists across the globe.
81 Pakistanis among world’s top 2% scientists | The Express Tribune
Eleven academic of QAU figure in list compiled by Stanford University researchers
- Messages
- 5,369
- Reactions
- Nation of residence
-
- Nation of origin
-
Chinese Tencent has published its progress on the research of mobile robot
- Messages
- 8,145
- Reactions
- Nation of residence
-
- Nation of origin
-
Your data and how it is used to gain your vote
By Jane WakefieldTechnology reporter
Published24 minutes ago
Related Topics
Targeted advertising is common, but the same thing is happening in politics with far less clarity
How much do political parties know about you - and how is it used to try to sway your vote?
The Cambridge Analytica scandal threw light on how the Facebook data of millions was harvested and turned into a messaging tool.
The revelations were criticised far and wide by politicians of all stripes.
But now, a report from the UK's Information Commissioner's Office (ICO) puts the spotlight on the relationship between data brokers and the politicians here.
Should we be concerned?
Even limited information can be used in surprising ways, the ICO report found.For example, buying someone's name can lead to making guesses about their income, number of children and ethnicity - which is then used to tailor a political message for them.
The report suggests that the Conservative Party is doing just that, using so-called "onomastic data": information derived from the study of people's names which could identify their ethnic origin or religion.
It has done that for 10 million voters, most of whom will be unaware of exactly how their information is being used.
Political parties can legitimately hold personal data on individuals to help them campaign more effectively. But sophisticated data analytics software can now combine information about individuals from multiple sources to find more about their voting characteristics and interests - something some people may find disturbing.
"Data collection is out of control and we need to put limits on what is collected," says Lucy Purdon from Privacy International (PI).
So how do the parties get my data in the first place?
The electoral register forms "the spine" of data sources, according to PI, but beyond that it is surprisingly difficult to work out what the parties use.What has become clearer in recent months is the role of data brokers. Both the Conservatives and the Labour Party make use of a product from Experian called Mosaic, according to the Open Rights Group (ORG), which describes Experian as being a "one-stop shop for data used in political profiling".
Experian is better known as a credit rating agency, but it also acts as a data broker, along with others such as Equifax and Transunion.
They collect data themselves or buy it from other companies, such as a credit card company. They also crawl the internet for useful information about people and aggregate that with data from other sources.
They then sell it on to advertisers – or, in this case, to political parties.
A two-year investigation by the ICO found that millions of adults in the UK had had their data processed by Experian. The ICO recommended a long list of improvements the company needed to make in order to comply with the EU-wide GDPR law on data privacy.
A PI complaint sparked the ICO investigation. PI says "it is a complex and opaque industry, and we are just starting to chip away at how this eco-system works".
How do political parties use your data?
image captionOur digital fingerprint can identify a huge amount about us
Having data on a person means that political messages can be personalised, and while this is a good way to hammer home specific messages, it could be argued that it is also giving people only part of the story about any given political issue.
According to PI it helps to create "echo chambers, polarise votes and restrict political debate".
"If someone has the given name Mohammed, for example, it may be inferred that they are from an immigrant family and so messages about immigration can be tailored," says Jim Killock from the ORG.
"Or if there are two people with the same surname living at an address, it can be guessed that they may be married and messaging tailored to that."
What do the political parties say?
The BBC has asked the Conservatives, Labour and the Liberal Democrats how they use data and where they receive it from. None have replied.The ORG conducted its own investigation and as part of its research it asked people to request all data political parties held on them, something known as a Data Subject Access Request. Few got responses but the scant information gleaned included:
- Labour had compiled up to 100 pages of data per individuals, broken down into over 80 categories
- Liberal Democrats attempted to guess the number of families in a home, and an individual's age based on name
- Conservatives attempted to estimate how likely an individual was to read and enjoy the Daily Mail, as well as guessing income
The Labour Party did not reply. The Conservatives said that they did purchase commercially available data, but did not say what they did with it.
Following the ICO revelations about onomastic data, the ORG has contacted the Conservative Party asking if it still uses this data. It has not yet had a response.
Much of the use of personal data by political parties is done under the banner of democratic engagement, which is used to justify a wide range of profiling activities.
What can be done about it?
The ICO says political parties need to be much clearer about how they intend to use personal data.But the Open Rights Group thinks it needs much tougher action.
"If it does not crack down, there is no incentive for better behaviour," it said.
One of the obvious ways would be to allow voters the ability to refuse the sharing of their data between a political party and a third party, such as a data broker.
GDPR stipulates that individuals should know exactly how their data is being used and agree to that.
But that could be harder because of how little is known about what data is being collected in the first place, PI's Ms Purdon said.
"The data broker industry is so complex and while the GDPR gave people more rights over their data, how are you supposed to exercise those rights if don't even know a company is collecting your data and profiling you?"
Your data and how it is used to gain your vote
How much is really known about the data bought by political parties, and how it is used?
www.bbc.com
