AI Outwits Goalkeepers by Anticipating Penalty Takers’ Shots

Goalkeepers have difficulty predicting the direction of penalty shots

Javier Soriano/AFP via Getty Images

Trained with a dataset of over 1,000 penalty kicks, the deep learning model demonstrates superior predictive capabilities compared to actual goalkeepers.

“Penalty kicks are often decisive moments in football that can influence the results of important tournaments,” states David Freire-Obregón from the University of Las Palmas de Gran Canaria, Spain. “Yet, real-time support for goalkeepers mostly relies on intuition. We sought to determine if machine learning could effectively predict the direction of a shot based on the kicker’s movements.”

Freire-Obregón and his team analyzed 1,010 penalty kicks from televised matches in Spain. Out of these, 640 were deemed usable by the AI, while the remainder were excluded due to blurriness, being too brief, or other obstructions.

Each video clip was processed through 22 different deep learning models. The goal was to predict whether the penalty kick would go left, right, or center left, or lower right, depending solely on the player’s body posture and footedness.

The top-performing model was able to accurately identify the direction of the ball—right, left, or center—52% of the time. Excluding the middle option improved the model’s accuracy to 64%.

The researchers were astonished, stating, “We can uncover the intent behind subtle movements, which serve as clues before the ball is kicked,” says Freire-Obregón. He hopes this insight will aid training for goalkeepers, although he admits utilizing AI predictions in actual matches presents additional challenges.

“Our next goal is to determine if we can predict the outcome of a penalty kick solely based on the kicker’s pre-shot movements,” he adds. “If feasible, how quickly can such predictions be made while retaining acceptable accuracy?”

topic:

Source: www.newscientist.com

Anticipating Nintendo’s Switch 2 Annoucement: Enhanced Power, Magnetic Controllers, and Backwards Compatibility

Nintendo may announce its next console this week, a successor to the Nintendo Switch, which was released in March 2017 and sold 150 million units. There’s just one problem. That said, we already know almost everything about it. There is little that Nintendo can announce at this point that will come as a surprise to anyone who has been following the rumors closely.

Nintendo Switch 2 leaks started trickling in last summer and escalated to a flood this month. Last week, CES Technology Trade Show In Las Vegas, accessory maker Genki arrived with a complete model of Nintendo’s next console, which they were happy to show off behind closed doors to explain their upcoming product. You can also see detailed renderings on Genki’s website. It’s a slightly larger and more powerful version of the Switch console we know and love, with controllers that attach magnetically to the side of the screen rather than sliding in and out. Play while docked to your TV or on the go.

This is a very un-Nintendo approach. Aside from the NES/SNES, all of Nintendo’s consoles ushered in a revolution in form factors. There was the N64, with its pioneering analog sticks and three-pronged controller. GameCube looks like a stubby toy. Wii, motion control remote control included. Its successor, the Wii U, added a screen in the center of the controller. With the exception of the dual-screen DS and its successor, the 3DS, which added stereoscopic 3D to the console’s capabilities, this is the first time Nintendo has produced two consecutive consoles that look and act the same. They even share a name and logo. The most reliable information currently indicates that it will be called Nintendo Switch 2.

I won’t repeat any more details that were leaked about the Switch 2. They are easy to search and within the next day or so you can clearly see what is true and what is false. Nintendo has confirmed that the Switch 2 will share its back catalog with the Switch. This will allow all players to enjoy all the games they have purchased over the past eight years on their new console. We also know it won’t be out until April (June is my money), as it’s scheduled to come out in Nintendo’s next fiscal year. However, this is an unusual situation. We know almost everything about the console from gaming’s most secretive company even before it’s officially announced. How did it happen?

It was difficult to get a PS5 on release day. Photo: Charlie Tribalew/AFP/Getty Images

When the PlayStation 5 was released in 2020, the biggest talking point at the time was that people wouldn’t be able to get their hands on the PlayStation 5. Some customers who pre-ordered the PS5 received a package containing a bag of rice instead, but it was swapped with a vendor who was having trouble with the delivery chain. On eBay and other resale platforms, consoles were selling for two to three times the retail price. The supply-demand gap has dogged gaming consoles for at least the first two years, caused in part by manufacturing challenges during the pandemic. Nintendo probably wanted to avoid a similar situation.

We know that Nintendo’s manufacturing partners have been manufacturing parts for this console for a long time, over a year. The company aims to maintain large amounts of inventory in preparation for product launches. This is one of the reasons why so much information was leaked in advance. Various companies are already involved in the production of the Switch 2, and units and some units have been out for quite some time.

Nintendo also hasn’t gone after leakers or legally shut anything down in the way you might expect. The company’s only response to this deluge of unauthorized information, given to Japan’s Sankei Shimbun last week, was that “these images and videos are not official.” This suggests that Nintendo itself thought this might be inevitable. The company is delaying the announcement of its next console for as long as possible to preserve the survival of the phenomenally successful Switch, and said it doesn’t think these leaks will significantly damage its sales outlook.

The Switch 2 announcement will likely contain some surprises. What’s surprising is the rather un-Nintendo nature of this iterative console, and the piecemeal nature of what we’re discovering about it. Stay tuned for official announcements coming soon for more details.

what to play

Literally mow the grass. It’s literally just mowing the grass. Photo: Protostar

Effortless dad games for those who don’t want to spend time in the garden for a quiet January: It’s literally just mowing the grass. That’s exactly right. With a swipe, the small riding lawn mower eases its way through the ever-widening swathes of rough grass in your neighbor’s yard until the entire street is tidy. Cut the grass, collect hats, tap and admire different types of butterflies. It was my friend Patrick Klepek from the pro-gamer newsletter who brought this to my attention. cross play (We do a podcast together about navigating games with kids), and I was surprised to find myself playing it for a full 30 minutes straight. Am I getting old?

Available: iOS/Android

Estimated play time: 5 minutes or 1 hour, as long as you like

what to read

Dreams on a Pillow took 10 years to create. Photo: Rasheed Abueide
  • Games about Dreams on a Pillow 1948 Nakba Palestinian developer Rasheed Abueideh has reached his fundraising goal. I spoke to Abueide about the many obstacles he faced in trying to tell the Palestinian story through video games, challenges that no one should have to face.

  • Square Enix announces new policy The purpose is to protect staff from: Harassment by toxic fansit goes beyond restricting games and services for players who abuse support staff and developers.

  • Latest Great game in no time speed running event Last weekend, we raised more than $2.5 million for charity. A personal highlight was the Crazy Taxi player accompanied by a live pop-punk band.

Skip past newsletter promotions

What to click

question block

Source: www.theguardian.com

The Mechanisms of Anticipating the Beat Drop in Your Brain

We are able to enjoy music because of our ability to recognize musical boundaries.

NDAB Creativity/Shutterstock

We may finally understand how the brain processes beat drops: People use two distinct brain networks to predict and identify the transitions between musical segments.

Musical boundaries – the moments when one part of a composition ends and another begins – are important to enjoying music, especially in the Western musical tradition. Without them, he says, your favorite hits can sound like a monotonous, random stream of notes, “like reading a text without punctuation.” Ibarra Burnat Perez At the University of Jyväskylä, Finland.

To understand how the brain processes musical boundaries, she and her colleagues analyzed brain activity while listening to 36 adults listen to instrumental pieces from three different genres: Adios Nonino Astor Piazzolla, an American progressive metal band Stream of consciousness Dream Theater and Russian Ballet Classics of Spring Festival Works by Igor Stravinsky. All of the listeners had attended school in Finland, and half of them considered themselves semi-professional or professional musicians.

The researchers found that just before musical boundaries, a brain network they call the early auditory network activates in anticipation of the end of a musical phrase. This network primarily involves auditory regions located in the posterior, or back, outer region of the brain called the cortex.

Another network becomes active during and after musical transitions. This network, called the border-transition network, is characterized by increased activity in auditory areas toward the middle and anterior, or front, parts of the cortex. Perez says that this change in brain activity between the two regions is similar to how the brain understands the difference between sentences in a language.

During and after the musical boundary, several brain regions, including the right ventrolateral prefrontal cortex, which is involved in complex cognitive tasks and decision-making, deactivate, suggesting that the brain redirects attention and resources to integrating new musical information as a new segment begins, Perez says.

Musicians and non-musicians also used these two brain networks differently. For example, musicians relied on brain regions important for higher-order auditory processing and integration, which may reflect a more specialized approach to understanding musical boundaries, Perez says. Non-musicians, on the other hand, showed greater connectivity across broader brain regions, indicating a more general approach.

In addition to shedding light on how the brain processes music, Perez says, these findings could also help develop music therapy for people who have difficulty comprehending language. For example, incorporating elements of musical boundaries into speech transitions (such as matching syllables to a melody) might make sentences easier to understand, she says.

topic:

Source: www.newscientist.com

Anticipating the Future: 8 AI Predictions for 2024

It was shocking for AI, as it moved from niche to mainstream technology faster than ever before. But 2024 will be the year when this hype really becomes reality as people consider the capabilities and limitations of AI as a whole. Here are some ways we think that could happen.

OpenAI becomes a product company

After a management shake-up in November, OpenAI will be a different company — it may not look like it on the outside, but the trickle-down effect of Sam Altman taking more full charge will make all the difference. You can feel it on the level. And one of the ways we expect that to manifest is through the idea of “shipping.” You can see that in the GPT store. Originally he was scheduled for release in December, but was understandably delayed due to executive turmoil. “AI app store” will continue to be strongly promoted as a platform to get AI toys and tools. Don’t worry about Hugging Face or any other open source model. They have a great model called Apple and they follow it all the way to the bank. Expect to see more similar moves from OpenAI in 2024, as the prudence and academic reserve exhibited by previous boards gives way to unseemly greed for markets and customers. Other big companies working on AI will likely follow this trend (e.g. we expect Gemini/Bard to get into a ton of Google products), but I suspect it will be more pronounced in this case.

Agents, generated videos, and generated music graduate from quaint to experimental

Some niche applications of AI models, such as agent-based models and generative multimedia, will grow beyond “meh” status by 2024. If AI is going to help you do more than summarize things or create lists, it will need access to spreadsheets, ticket-buying interfaces, transit apps, and more. In 2023, several attempts were made with this “agent” approach, but none really caught on. I don’t expect anything to really take off in 2024 either, but I think agent-based models will look a little more convincing than they did last year. We’ll also see some clutch use cases that are notorious for tedious processes such as submission. Insurance claims. Video and audio will also find a niche where their shortcomings are less obvious. In the hands of skilled creators, the lack of photorealism will not be an issue and AI video will be used in fun and interesting ways. Similarly, generative music models are likely to be adopted by some major productions, such as games, where professional musicians can also leverage the tools to create endless soundtracks.

The limitations of monolithic LLM become clearer

So far, there’s been a lot of optimism about the capabilities of large-scale language models, and they’ve actually proven to be better than anyone expected, and can be used to create more computing power. As more are added, its capabilities further increase accordingly. But 2024 will be the year that something will be given. There is a lot of research going on at the forefront of this field, so it is impossible to predict exactly where this will happen. The seemingly magical “new” features of the LLM will be further studied and understood in 2024. Also, things like LLM not being able to multiply large numbers would make more sense. At the same time, the returns on number of parameters also start to decrease, and while training a 500 billion parameter model might technically yield better results, the compute required to do so could probably be more effectively deployed. may be. A single monolithic model is unwieldy and expensive, whereas a combination of multiple experts (a collection of smaller, more specific and perhaps multimodal models) is much easier to update in parts. However, it may prove to be nearly as effective.

Marketing meets reality

The simple fact is that it will be very difficult for companies to live up to the hype built in 2023. Marketing claims about machine learning systems that companies deploy to keep up will be subject to quarterly and annual reviews…and there is a strong possibility that they will be found to be inadequate. It is high. We expect significant customer withdrawal from AI tools as the benefits do not justify the costs and risks. At the other end of the spectrum, we may see litigation and regulatory action with AI service providers who are unable to substantiate their claims. Capabilities will continue to grow and advance, but it is not far off that all 2023 products will survive, and there will be a round of consolidation as the wave’s erratic riders decline and are consumed.

Source: techcrunch.com