Send Link mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
https://mailman.anu.edu.au/mailman/listinfo/link
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Link digest..."
Today's Topics:
1. Re: New Chinese Silicon Photonics (Antony Barry)
2. Sports AI video analysis and predictive analytics
(Stephen Loosley)
3. Antimatter-sniffing detector to monitor nuclear reactors on
Earth etc (Stephen Loosley)
----------------------------------------------------------------------
Message: 1
Date: Mon, 7 Oct 2024 15:49:17 +1100
From: Antony Barry <[email protected]>
To: Stephen Loosley <[email protected]>
Cc: link <[email protected]>
Subject: Re: [LINK] New Chinese Silicon Photonics
Message-ID:
<CAECOtWy1Os=_teqjfjgup4whc_bcl26fwv9jvcavrvgz2wh...@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"
Here's a summary of the article written by Perplexity AI:
A state-funded semiconductor lab in Wuhan, China, called JFS Laboratory,
has announced a breakthrough in silicon photonics technology. The lab
successfully integrated a laser light source with a silicon-based chip,
marking a first for China in this field.Key points:
1. Silicon photonics uses optical signals instead of electric signals
for transmission, potentially overcoming current limitations in chip design.
2. This achievement could help China progress towards semiconductor
self-sufficiency amid US sanctions.
3. Major global semiconductor companies, including TSMC, Nvidia, Intel,
and Huawei, are also working on silicon photonics technology.
4. The global market for silicon photonics chips is expected to grow
significantly, reaching US$7.86 billion by 2030.
5. Silicon photonics may be particularly advantageous for China, as it
can be produced using relatively mature materials and equipment without
relying on advanced EUV lithography machines, which are subject to export
controls.
6. This development could potentially shift the focus of US-China tech
competition towards emerging technologies like silicon photonics.
The breakthrough is seen as a significant step for China in addressing
technical barriers in chip design and reducing reliance on foreign
technologies.
On Sun, Oct 6, 2024 at 5:14?PM Stephen Loosley <[email protected]>
wrote:
>
> China claims breakthrough in silicon photonics that could clear technical
> hurdle
>
> A Wuhan-based lab has announced a milestone that could help China overcome
> restraints imposed by traditional chip-design technology
>
> By Xinmei Shen 6 Oct 2024
> https://www.scmp.com/tech/tech-war/article/3281156/chip-war-china-claims-breakthrough-silicon-photonics-could-clear-technical-hurdle
>
> [Photo caption: A state-fund ed lab in China has announced a breakthrough
> in chip-design technology. Photo: Shutterstock]
>
>
> A state-funded semiconductor lab in China said it has achieved a milestone
> in the development of silicon photonics, which could help the country
> overcome current technical barriers in chip design and achieve
> self-sufficiency amid US sanctions.
>
> JFS Laboratory, based in Wuhan, capital of central Hubei province and a
> national base for photonics research was able to light up a laser light
> source integrated with a silicon-based chip.
>
> This is the first time that this has been successfully done in China,
> according to a blog post published by the lab last week.
>
> The achievement means that China has filled one of the few blanks in its
> optoelectronics technology, state media Peoples Daily reported on Friday.
>
> Silicon photonics rely on optical signals instead of electric signals for
> transmission. It aims to address the restraints imposed by current
> technology, as the transmission of electric signals between chips is
> approaching its physical limit, the lab said.
>
> Established in 2021 with 8.2 billion yuan (US$1.2 billion) in government
> funding, JFS is one of Chinas key institutions tasked with pursuing
> technological breakthroughs.
>
> Major players in the global semiconductor industry have devoted resources
> into advancing silicon photonics, which is believed to hold the future to
> making better chips for data and graphics processing, as well as artificial
> intelligence (AI).
>
> Still, businesses have faced challenges in translating scientific
> breakthroughs into commercial products.
>
> Taiwan Semiconductor Manufacturing Company, a world top contract-chip
> maker, is one of the companies working on the technology. Its
> vice-president, Douglas Yu Chen-hua, last year said that a good silicon
> photonics integration system could address critical issues in energy
> efficiency and computing power in the AI era.
>
> That development would bring about a paradigm shift in the industry, he
> said.
>
> US chip design giants Nvidia and Intel, as well as Chinas Huawei
> Technologies, are also eyeing advances in silicon photonics. The global
> market for silicon photonics chips is expected to reach US$7.86 billion by
> 2030, up from US$1.26 billion in 2022, according to estimates by SEMI, an
> international semiconductor industry association.
>
> [Photo caption: Companies such as Taiwan Semiconductor Manufacturing
> Company see silicon photonics as the future of chip design. Photo: Reuters]
>
> Silicon photonics may present an even bigger opportunity in China, where
> US export controls on advanced chip-making technologies have hindered the
> development of traditional semiconductors.
>
> Silicon photonics chips can be produced domestically using relatively
> mature raw materials and equipment without relying on high-end extreme
> ultraviolet (EUV) lithography machines, unlike electrical chips, Sui Jun,
> president of Beijing-based semiconductor start-up Sintone, was quoted as
> saying by local media in 2022.
>
> EUV machines, required for making advanced chips, are considered the
> Achilles heel of the Chinese semiconductor industry, as domestic firms
> struggle to mass-produce such tools. Netherlands-based ASML, which holds a
> virtual monopoly on EUV machines, stopped exporting the equipment to China
> in 2019.
>
> Silicon photonics could become an emerging front in US-China tech
> competition, according to a report published by US think tank Centre for
> Strategic and International Studies (CSIS) in January.
>
> "While the US-led export controls are likely setting back Chinese
> capabilities in the manufacture of traditional chips ? [they] could also
> inadvertently incentivise China to devote more resources to emerging
> technologies that will play an important role in next-generation
> semiconductors,? Matthew Reynolds, a former economics programme fellow at
> the CSIS, wrote in the report.
>
>
>
> Xinmei Shen joined the Post in 2017 and is a technology reporter. She
> covers content, entertainment, social media and internet culture.
> Previously, she was with the Post?s tech news site, Abacus. Before that,
> she was a reporting intern at The Information whilst studying at the Univ
> --
>
>
> _______________________________________________
> Link mailing list
> [email protected]
> https://mailman.anu.edu.au/mailman/listinfo/link
>
--
Mob:04 3365 2400 Email: [email protected], [email protected]
------------------------------
Message: 2
Date: Mon, 7 Oct 2024 06:01:18 +0000
From: Stephen Loosley <[email protected]>
To: link <[email protected]>
Subject: [LINK] Sports AI video analysis and predictive analytics
Message-ID:
<me0p282mb4413e6a660e0e83259d10eadc2...@me0p282mb4413.ausp282.prod.outlook.com>
Content-Type: text/plain; charset="Windows-1252"
Please Note: This email did not come from ANU, Be careful of any request to buy
gift cards or other items for senders outside of ANU. Learn why this is
important.
https://www.scamwatch.gov.au/types-of-scams/email-scams#toc-warning-signs-it-might-be-a-scam
Youth Sports Were Already Intense?Now AI Tools Are Supercharging the Competition
Clubs and high school teams increasingly have access to AI video analysis and
predictive analytics that are professionalizing youth sports.
By Todd Feathers Published October 4, 2024
https://gizmodo.com/youth-sports-were-already-intense-now-ai-tools-are-supercharging-the-competition-2000506254
Ashley Brown used to watch her daughter?s club volleyball games through the
screen of her phone, afraid to put it down and miss out on footage of a set or
kill that would catch the eye of a college recruiter.
And as the coach of her daughter?s high school volleyball team in Caledonia,
Michigan, Brown?s attention was constantly split between watching the games and
tallying each player?s statistics by hand.
But this year, her daughter?s traveling club team purchased an artificial
intelligence service called Balltime for all players aged 12 to 18.
A single phone or tablet placed behind the court?s endline records a game and
uploads it to the company?s platform, which uses body and object recognition
algorithms to track each player so that their every ball contact and movement
on the court can be cataloged and datafied.
By the time a player has gotten home from a game and showered, the service can
prepare personalized stat reports and social media-ready highlight packages.
It also gives coaches a bevy of data that was previously only available to
professional and elite college volleyball programs.
Balltime automatically measures how high in the air players make contact with
the ball, their kill and error percentages, ball trajectories, serve speeds,
and which rotations of players score the most.
Watch: Uber is Getting in the Blue Check Mark Game
It?s part of a growing sports technology industry selling computer vision
algorithms, wearable biometric sensors, and predictive analytics services to
youth clubs and high school athletics departments, opening up a new world of
video and data analysis that?for better or worse?is changing the way young
athletes and their families experience sports.
Without needing to spend hours cutting together videos themselves, teams can
gather comprehensive video evidence to show, rather than tell, young players
what they did right and wrong.
And coaches and college recruiters say platforms like Balltime and Darkhorse
AI, which provides a similar player-tracking service for soccer, are allowing
them to make more data-driven decisions about rosters and playing time.
?It has helped me already this season with some of the difficult conversations
I?ve had to have with players and parents,? Brown said.
?[I can tell them] it?s not because I don?t like your kid, this is a computer
system and software system that are rating these things based on these
parameters.?
As valuable as they can be to help players learn and improve, some coaches
worry the data and highlight packages produced by AI analytics services also
supercharge unhealthy competition among young athletes vying for attention from
recruiters and on social media.
?There?s this mad rush right now to use these tools to self promote and there?s
this pit of loneliness that can happen when you don?t get that attention,? said
Ben Bahr, a former college coach and data analyst who now works as director of
coaching for Adrenaline Volleyball in Iowa, which uses Balltime.
?With the rise of AI and sharing of data, the thing that?s come out of this the
most is that it?s become much easier to compare yourself with what someone else
is doing.?
Volleyball Ai
Big money in young sports
The push for advanced data analytics is part of the growing monetization of
children?s sports.
A report frequently cited by sports technology investors estimates that the
youth sports market had a global value of $37.5 billion in 2022 that will grow
to $69.4 billion by 2030, rivaling some of the world?s most popular
professional leagues.
Private equity firms have spent hundreds of millions of dollars to buy youth
sports complexes and teenagers are now competing not just for spots on college
rosters but also for life-changing money from name, image, and likeness (NIL)
deals, thanks to a 2021 Supreme Court case that opened the door to private
sponsorships for college athletes.
?There is definitely a downward move toward more professionalized youth
sports,? Dan Banon, Balltime?s CEO, told Gizmodo. He and chief technology
officer Tom Raz began building the platform with adults in mind but soon
realized the biggest potential for growth was in traveling club teams and high
schools.
Over the past year, he said, the company has seen more signups from junior
varsity teams and even middle school programs.
Their data shows that some players spend seven hours a month reviewing footage
on Balltime.
At $25 per month for a player, Balltime?s recruiting package isn?t for everyone.
But with the average household spending $883 a year on a single child?s primary
sport, according to a parent survey from the Aspen Institute?s Project Play,
the additional cost is also well within many families? sports budgets.
Responding to pressure from families who want their children to have every
advantage, some elite clubs are looking for even more ways to combine, collect,
and analyze player data.
Mustang Soccer League, based in Danville, California is in the process of
building out a data analytics department and some players can expect to spend
an additional $250 a year on technology subscriptions, said Fred Wilson, the
club?s executive director.
Mustang recently introduced Darkhorse AI for its 12 to 18-year-old teams and is
beginning to discuss high-level analytics with players as young as 10.
Like Balltime, Darkhorse uses object recognition algorithms to track players
during games, automatically cataloging various stats and curating highlight
reels.
Some Mustang teams also link that information with biometric data like heart
rate and running speed captured by Beyond Pulse wearable sensors.
?I don?t know how much learning we?re going to do with 10-year-olds, but I?m
trying to instill a habit so that when they?re 14 or 15 they?re paying
attention to these things ? make it just second nature for players to
understand,? Wilson said.
The club boasts several former players who are now professionals and dozens
more at top college programs.
?This whole AI piece takes us to that next level to be able to do that,? Wilson
said.
?I take 200 calls a year [from sports technology vendors] to find the gem.?
Most of them are ?out trying to make a quick buck,? he added, but some are
offering real value.
Karin Pfeiffer, director of the Institute for the Study of Youth Sports at
Michigan State University, said that even at the college level, where wearable
biometric sensors and data analytics have been common for some time, programs
are still struggling to figure out what data is actually useful for athletes
and coaches.
?Collegiate level coaches are approached all the time with these technology
things, I imagine that?s going to bleed down to high school too if it hasn?t
already,? she said.
?You can get so much information out of it, but the question is what?s
relevant, what?s actually tied to performance, what?s tied to future success.?
?Insane pressure? to hit metrics
Coaches and company executives told Gizmodo that the biggest driver of the AI
analytics boom in youth sports is the prospect that the tools can help athletes
make the jump to college, where a spot on a roster can translate into tens of
thousands of dollars in scholarships and, at the highest levels, hundreds of
thousands of dollars in sponsorship deals.
Some products, like SwimIntel, focus purely on recruiting rather than coaching.
The platform collects competition data on swimmers as young as 15 and uses it
to rank them as recruits and train models that predict how they will fare in
different collegiate swim programs. For $40 a month, swimmers can receive
60-plus page analytics reports that project how their times will improve, or
worsen, at different colleges. Schools that contract with SwimIntel receive
similar predictions in reverse based on how other athletes from the same youth
swim club have performed once in college.
?We let college coaches play moneyball using AI,? said Jamie Bailey, the
founder of SwimIntel. ?We let student athletes use AI to find best fits. And in
the end, what we?re trying to do is reduce that dropout rate. One out of six
college swimmers don?t come back their sophomore year.?
Bahr, the former college volleyball coach and data analyst, said that when he
worked at programs like Baylor University and Southern Methodist University the
volleyball staff would sometimes receive 600 emails a day from prospective
recruits.
If a player didn?t catch the recruiter?s eye in the first 30 seconds of their
highlight reel, they were often passed over.
Now with Balltime, more players have access to more video footage than they
ever had before and the measurement algorithms have changed the way college
programs assess highlight reels.
?I don?t even need to watch the film,? Bahr said, recruiters can just look at
Balltime?s video analytics to see ?are you touching the ball at a height that?s
at a better height than our competition? Are you touching the ball over 10 feet
or not? We?re already in this insane pressure of hitting these metrics and
these tools certainly haven?t helped with that?
At the same time, several coaches who spoke to Gizmodo were optimistic that AI
video tools will also increase competition in a positive way?by allowing
athletes who don?t play for the biggest youth teams to improve their skills and
get noticed by recruiters.
?Having that video, having those stats, those are real educational tools,? said
Pfeiffer, from Michigan State University.
?It all comes down to how the athlete is receiving that and making sure
appropriate supports are in place. I don?t think these things should be
unchecked, they should come with guidance from parents and coaches.
But sometimes parents and coaches are overzealous?
---
------------------------------
Message: 3
Date: Mon, 7 Oct 2024 06:11:12 +0000
From: Stephen Loosley <[email protected]>
To: link <[email protected]>
Subject: [LINK] Antimatter-sniffing detector to monitor nuclear
reactors on Earth etc
Message-ID:
<me0p282mb44131e9c2699e5c4a5814a1cc2...@me0p282mb4413.ausp282.prod.outlook.com>
Content-Type: text/plain; charset="Windows-1252"
Antimatter Could Be the Key to Solving Big Mysteries
Two new studies highlight the enigmatic nature of antimatter, revealing its
potential role in both understanding the universe's origins and unlocking the
secrets of particle physics.
By Isaac Schultz Published October 4, 2024
https://gizmodo.com/scientists-are-using-antimatter-to-unlock-the-universes-biggest-secrets-2000506504
Two papers published this week showcase the perplexing origins and potential
uses of antimatter, a type of matter that flips the rules governing ordinary
matter onto their heads.
One paper, published today in JCAP, found that antinuclei from cosmic rays may
be an indicator of a specific kind of a dark matter.
In a separate paper, published earlier this week in AIP Advances, researchers
describe a method of detecting nuclear reactors? locations and activity using
antineutrinos produced by the facilities? nuclear reactions.
Antimatter is important because it may help to explain fundamental cosmic
mysteries, like why the universe is made of matter instead of an equal mix of
matter and antimatter. These studies fit into a larger effort to crack some of
physics? biggest puzzles, including the nature of dark matter, physics at the
smallest scales, and possibly even the origin of the universe itself.
Despite its name, antimatter is literally matter. It has mass. Antimatter
refers to a group of particles that have opposite electrical charges to their
ordinary counterparts. You?ve heard of electrons (which have a negative charge)
and protons (with a positive charge); their antimatter counterparts are
positrons (with a positive charge) and an antiproton (a negative charge).
Though there are differences in the charge of the particles, antimatter isn?t
entirely alien to the fundamental forces. Last year, a team of physicists found
that antimatter reacts to gravity the same way as ordinary matter, a finding
that affirmed both Einstein and the Standard Model of Particle Physics.
Something more similar to the idea of ?antimatter? you may have in your head is
dark matter?which also has mass?but is invisible to every kind of detector
humankind has so far devised. Scientists know dark matter exists because its
gravitational effects are visible, even though the particle (or particles!)
responsible cannot be directly observed.
Antimatter remains a matter of confusion (sorry, awful pun) for a few reasons.
As explained by Gizmodo in 2022:
The universe rocked into being 14 billion years ago, with a Big Bang that in
theory should have created equal amounts of matter and antimatter. But look
around you, or at the latest Webb telescope images: We live in a universe
dominated by matter. An outstanding question in physics is what happened to all
the antimatter.
Antimatter and dark matter dovetail neatly in the recent JCAP paper, which
posits that the amount of antimatter detected by experiments is more than there
should be?and they believe dark matter is the culprit.
A few different particles (and other, more exotic objects) have been posited as
responsible for dark matter. Among them: axions, a particle named for a laundry
detergent; MAssive Compact Halo Objects, or MACHOs; dark photons, which despite
their name are more like axions than some insidious version of light; and
primordial black holes, which would be minuscule black holes birthed at the
beginning of the universe, floating through space.
The recent research focuses on another type?Weakly Interacting Massive
Particles, or WIMPs?as the guilty party. The theory is essentially that when
WIMPs collide, they sometimes annihilate?destroy one another?emitting energy
and particles of matter and antimatter.
In the aforementioned 2022 research, a team of physicists using the ALICE
experiment at CERN found that antimatter could travel through our galaxy with
ease instead of being snuffed out by the matter in the interstellar medium, a
redeeming conclusion for antinuclei detectors like the AMS-02 experiment aboard
the International Space Station.
?Theoretical predictions suggested that, even though cosmic rays can produce
antiparticles through interactions with gas in the interstellar medium, the
amount of antinuclei, especially antihelium, should be extremely low,? said
Pedro De la Torre Luque, a physicist at the Institute of Theoretical Physicists
in Madrid and lead author of the JCAP paper, in a SISSA Medialab release.
?We expected to detect one antihelium event every few tens of years, but the
around ten antihelium events observed by AMS-02 are many orders of magnitude
higher than the predictions based on standard cosmic-ray interactions,? De la
Torre Luque added. ?That?s why these antinuclei are a plausible clue to WIMP
annihilation.?
However, De la Torre Luque added that WIMPs could only explain the amount of
antihelium-3?one antimatter isotope detected by AMS-02?and not detected amounts
of the rarer, heavier antihelium-4. In other words, even if WIMPs are
responsible for dark matter, they don?t tell the whole story.
WIMPs could be responsible for the antimatter detections that space-based
detectors are collecting.
But regardless of the dark matter question?one that will take a long time to
answer?the design of an antimatter-sniffing detector to monitor nuclear
reactors on Earth shows practical applications in the here and now.
Together, these findings on antimatter could offer new ways to harness the
strange properties of the universe for practical use, while also helping us
better understand both the cosmos and our own planet.
--
------------------------------
Subject: Digest Footer
_______________________________________________
Link mailing list
[email protected]
https://mailman.anu.edu.au/mailman/listinfo/link
------------------------------
End of Link Digest, Vol 383, Issue 9
************************************
