Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Learning Physically Simulated Tennis Skills from Broadcast Videos demonstrate that system produces controllers for physically-simulated tennis players that can hit the incoming ball to target positions accurately using a diverse array of strokes (serves, forehands, and backhands), spins (topspins and slices), and playing styles (one/two-handed backhands, left/right-handed play). Overall,... show more

AK

524,027 subscribers

125,559 views • 3 years ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

10 Comments

Mike Rodriguez3 years ago

@Scobleizer This is amazing. That being said, I can imagine broadcast sports and media companies not taking kindly to this. They have the power to shut things down for a lot of us. As opposed to artists who might not have the budget to attempt to retain usage rights.

尬聊3 years ago

I know I'm too late but prepare for all the "actually their technique sucks" takes

James Darpinian3 years ago

Those movements don't look realistic at all. They're supposed to be physically simulated?

Noah Basketball2 years ago

Train your shooters on the same technology the top coaches use for player evaluation. #MakeMoreShots #WinMoreGames

Gustavo3 years ago

bloody incredible. Not lifelike just yet but not far off

William Lamkin3 years ago

I’d be curious to see this kind of research for MMA. I would imagine the system might come across interesting and novel submission moves, unexpected fake outs, discover unexploited weaknesses and strange ways of leveraging the human form for combat, etc.

Michael Ehrmantraut3 years ago

Two-handed backhand needs some work. Also forehands could use the same adjustment. Namely to take the racquet back higher and then drop it down into place (rather than the low take back).

Jama Noor3 years ago

I think this would allow for training in VR, where previously unscalable, now any significant set of high quality video should allow for improved human training. Everyone can afford a 1-on-1 tutor at the highest level in any sport.

Chloro3 years ago

@Algon_33

Akshay 🚀3 years ago

Interesting! I particularly liked the single handed backhand & the kick serve!!

Related Videos

Roger Federer was asked why the one-handed backhand is dissappearing from tennis. Starts by saying “I don’t know” then gives an incredible four-part answer: ▫️Modern coaches grew up playing two-handed backhand and passed it down to students. ▫️One-handed backhands take a lot of time to perfect control and coaches/players don’t want to dedicate the time when they can make faster improvements elsewhere. ▫️Tennis went from 50/50 between baseline and net play to 90/10 baseline play (two-handed backhands much more useful for defensive and sliding baseline play). ▫️Equipment for tennis has changed (stiffer strings, modern rackets, heavier balls) and slower surfaces…more difficult to attack with only one hand. *** Tennis Channel link:

Roger Federer was asked why the one-handed backhand is dissappearing from tennis. Starts by saying “I don’t know” then gives an incredible four-part answer: ▫️Modern coaches grew up playing two-handed backhand and passed it down to students. ▫️One-handed backhands take a lot of time to perfect control and coaches/players don’t want to dedicate the time when they can make faster improvements elsewhere. ▫️Tennis went from 50/50 between baseline and net play to 90/10 baseline play (two-handed backhands much more useful for defensive and sliding baseline play). ▫️Equipment for tennis has changed (stiffer strings, modern rackets, heavier balls) and slower surfaces…more difficult to attack with only one hand. *** Tennis Channel link:

Trung Phan

671,724 views • 1 month ago

Synchronize Dual Hands for Physics-Based Dexterous Guitar Playing discuss: We present a novel approach to synthesize dexterous motions for physically simulated hands in tasks that require coordination between the control of two hands with high temporal precision. Instead of directly learning a joint policy to control two hands, our approach performs bimanual control through cooperative learning where each hand is treated as an individual agent. The individual policies for each hand are first trained separately, and then synchronized through latent space manipulation in a centralized environment to serve as a joint policy for two-hand control. By doing so, we avoid directly performing policy learning in the joint state-action space of two hands with higher dimensions, greatly improving the overall training efficiency. We demonstrate the effectiveness of our proposed approach in the challenging guitar-playing task. The virtual guitarist trained by our approach can synthesize motions from unstructured reference data of general guitar-playing practice motions, and accurately play diverse rhythms with complex chord pressing and string picking patterns based on the input guitar tabs that do not exist in the references. Along with this paper, we provide the motion capture data that we collected as the reference for policy training.

Synchronize Dual Hands for Physics-Based Dexterous Guitar Playing discuss: We present a novel approach to synthesize dexterous motions for physically simulated hands in tasks that require coordination between the control of two hands with high temporal precision. Instead of directly learning a joint policy to control two hands, our approach performs bimanual control through cooperative learning where each hand is treated as an individual agent. The individual policies for each hand are first trained separately, and then synchronized through latent space manipulation in a centralized environment to serve as a joint policy for two-hand control. By doing so, we avoid directly performing policy learning in the joint state-action space of two hands with higher dimensions, greatly improving the overall training efficiency. We demonstrate the effectiveness of our proposed approach in the challenging guitar-playing task. The virtual guitarist trained by our approach can synthesize motions from unstructured reference data of general guitar-playing practice motions, and accurately play diverse rhythms with complex chord pressing and string picking patterns based on the input guitar tabs that do not exist in the references. Along with this paper, we provide the motion capture data that we collected as the reference for policy training.

AK

26,917 views • 2 years ago

I am so fortunate to still play tennis at my age. But there sure is a lot of (completely worth it) prep work involved! Here's some background on my pre-tennis routine from this video: 1. Put on two braces: one on each knee 2. Switch to my tennis glasses, and l always carry a backup pair with me, just in case. One pair is a throwback to my favorite pair from the 1970s 3. Wear gloves, both to protect my hands from sun, and also to help with carpal tunnel 4. Use exercise bands to warm up shoulders, triceps, and biceps for forehands/ backhands. 60 reps for each 5. Perform 20 squats + stretching. And an important tip: Watch the ball all the way to the contact point. At the pro level, 50% of balls are missed because of not watching the ball all the way to the contact point. At the rec level, it's 90%. Get out there and play!

I am so fortunate to still play tennis at my age. But there sure is a lot of (completely worth it) prep work involved! Here's some background on my pre-tennis routine from this video: 1. Put on two braces: one on each knee 2. Switch to my tennis glasses, and l always carry a backup pair with me, just in case. One pair is a throwback to my favorite pair from the 1970s 3. Wear gloves, both to protect my hands from sun, and also to help with carpal tunnel 4. Use exercise bands to warm up shoulders, triceps, and biceps for forehands/ backhands. 60 reps for each 5. Perform 20 squats + stretching. And an important tip: Watch the ball all the way to the contact point. At the pro level, 50% of balls are missed because of not watching the ball all the way to the contact point. At the rec level, it's 90%. Get out there and play!

Billie Jean King

47,394 views • 2 years ago

NEW: LUMISTAR built a tennis ball machine you can control from your wrist. TERO is a portable trainer that fires full-court shots, lets you call drills by voice, and gives solo players a way to practice without a coach or hitting partner. -Fires balls up to ~90 mph -Reaches about 85 feet across the court -Targets 35 shot positions for forehands, backhands, volleys, and movement drills -Wristband voice control lets you start, stop, or change drills without walking back -App control lets you adjust speed, spin, elevation, and training sessions -Delivers up to 4 hours of battery life Priced at $1,099.

NEW: LUMISTAR built a tennis ball machine you can control from your wrist. TERO is a portable trainer that fires full-court shots, lets you call drills by voice, and gives solo players a way to practice without a coach or hitting partner. -Fires balls up to ~90 mph -Reaches about 85 feet across the court -Targets 35 shot positions for forehands, backhands, volleys, and movement drills -Wristband voice control lets you start, stop, or change drills without walking back -App control lets you adjust speed, spin, elevation, and training sessions -Delivers up to 4 hours of battery life Priced at $1,099.

Ritwik Pavan

591,759 views • 1 month ago

In our league the play stops when the ball gets back to the pitcher. So getting it back in fast is everything! Our team was struggling with it so I came up with this game on the fly. One practice. Fixed. And the kids were BEGGING to play it again! Here's how it works: ⚾ Two teams. Defense takes their positions. ⚾ Hit through the full order with a tennis racket and tennis ball. ⚾ Every base touched before the ball gets back to the pitcher = 1 point. ⚾ For example: Pitcher secures the ball between second and third? 2 points. ⚾ Out at first before runner gets there or fly ball caught? Zero points. ⚾ Score and keep running if the ball still isn't back. The play never stops until the pitcher has it in the pitchers circle . Works on getting the ball back in fast, backing up throws, base running, communication, and teamwork. We saw a massive difference the very next game! PLEASE SHARE with someone this can help! ⚾💪🔥

In our league the play stops when the ball gets back to the pitcher. So getting it back in fast is everything! Our team was struggling with it so I came up with this game on the fly. One practice. Fixed. And the kids were BEGGING to play it again! Here's how it works: ⚾ Two teams. Defense takes their positions. ⚾ Hit through the full order with a tennis racket and tennis ball. ⚾ Every base touched before the ball gets back to the pitcher = 1 point. ⚾ For example: Pitcher secures the ball between second and third? 2 points. ⚾ Out at first before runner gets there or fly ball caught? Zero points. ⚾ Score and keep running if the ball still isn't back. The play never stops until the pitcher has it in the pitchers circle . Works on getting the ball back in fast, backing up throws, base running, communication, and teamwork. We saw a massive difference the very next game! PLEASE SHARE with someone this can help! ⚾💪🔥

David Klein

44,156 views • 6 months ago

Rich Paul on Klay Thompson and the narrative that the Heat need more playmaking: “Klay, next to Giannis, who’s going to draw two and kick. You have to consider the Heat’s system. The Heat have a system and guys play better in that system. It’s the right move for Klay because he don’t have to do as much. With Spo it’s free flowing, ball moves, they don’t depended on somebody playing one on one basketball. It’s very efficient basketball when you watch. That’s why guys tend to play better because you’re doing more while actually doing less. Guys look better in the Heat system. You’re playing with Giannis now who’s getting downhill and drawing two. Klay just has to make shots” (Via Game Over)

Rich Paul on Klay Thompson and the narrative that the Heat need more playmaking: “Klay, next to Giannis, who’s going to draw two and kick. You have to consider the Heat’s system. The Heat have a system and guys play better in that system. It’s the right move for Klay because he don’t have to do as much. With Spo it’s free flowing, ball moves, they don’t depended on somebody playing one on one basketball. It’s very efficient basketball when you watch. That’s why guys tend to play better because you’re doing more while actually doing less. Guys look better in the Heat system. You’re playing with Giannis now who’s getting downhill and drawing two. Klay just has to make shots” (Via Game Over)

Heat Central

99,111 views • 1 month ago

“I can always be better than I was yesterday”: Novak Djokovic is one of the greatest athletes of all time, but he told Gayle King that he’s aware of how his body responds differently now that he’s been playing elite tennis for more than two decades. Djokovic shared about scaling back on the court to spend more time with his family, falling in love with tennis as a young child in Serbia and Serena Williams’ return: “It’s a gift for us.” “Novak Djokovic: The Wolf in Winter,” a documentary on Amazon Prime about the tennis star, premieres next month.

“I can always be better than I was yesterday”: Novak Djokovic is one of the greatest athletes of all time, but he told Gayle King that he’s aware of how his body responds differently now that he’s been playing elite tennis for more than two decades. Djokovic shared about scaling back on the court to spend more time with his family, falling in love with tennis as a young child in Serbia and Serena Williams’ return: “It’s a gift for us.” “Novak Djokovic: The Wolf in Winter,” a documentary on Amazon Prime about the tennis star, premieres next month.

CBS Mornings

22,019 views • 2 months ago

Neurophysiologist Louisa Nicola trains NBA players' brains and studies Alzheimer's for a living, and she says this 5-minute drill can sharpen your brain: "Using just a tennis ball and an eye patch, you can train your cognitive reserve, your processing speed, and your reaction time." "Start by throwing the ball against the wall, right arm to right arm, then alternate hands. That alone engages almost all of your executive functions." "Then put on an eye patch to block out 50% of your vision, so your visual system has to work under load and stress. To make it harder, stand on one leg and bring your cerebellum and balance into it." "This is the same drill I used with NBA players to sharpen their reaction time." PS. We post daily breakdowns on brain health and longevity, so if that's you, follow us LongevityLab for more. Like and repost if you found value in this post:

Neurophysiologist Louisa Nicola trains NBA players' brains and studies Alzheimer's for a living, and she says this 5-minute drill can sharpen your brain: "Using just a tennis ball and an eye patch, you can train your cognitive reserve, your processing speed, and your reaction time." "Start by throwing the ball against the wall, right arm to right arm, then alternate hands. That alone engages almost all of your executive functions." "Then put on an eye patch to block out 50% of your vision, so your visual system has to work under load and stress. To make it harder, stand on one leg and bring your cerebellum and balance into it." "This is the same drill I used with NBA players to sharpen their reaction time." PS. We post daily breakdowns on brain health and longevity, so if that's you, follow us LongevityLab for more. Like and repost if you found value in this post:

LongevityLab

640,549 views • 1 month ago

Can vision-language-action (VLA) models generalize to diverse OOD tasks and align with customized objectives? 🤔 🚀 We introduce GRAPE, a plug-and-play algorithm to generalize robot policies via preference alignment. GRAPE unfolds three benefits to boost the generalizability of VLAs: 👉1. GRAPE aligns VLAs on a trajectory level and endows the model with the ability for global decision-making, instead of merely cloning behavior; 👉2. GRAPE implicitly models reward from both successful and failed trials to boost generalizability to diverse tasks; 👉3. GRAPE adopts a scalable preference synthesis algorithm to rank trajectories with preferences that align with arbitrary objectives. Our experiments on a diverse array of real-world and simulated robotic tasks reveal: 1⃣GRAPE enhances the performance of state-of-the-art VLA models, increasing success rates on in-domain and unseen manipulation tasks by 51.79% and 60.36%; 2⃣GRAPE is versatile to be aligned with diverse objectives and reduce collision rates by 44.31% or rollout length by 11.15% when aligning towards safer or more efficient manipulation policy, respectively. Check out our full project for more details: 🔥 Paper: 🔥 Project: 🔥 Code:

Can vision-language-action (VLA) models generalize to diverse OOD tasks and align with customized objectives? 🤔 🚀 We introduce GRAPE, a plug-and-play algorithm to generalize robot policies via preference alignment. GRAPE unfolds three benefits to boost the generalizability of VLAs: 👉1. GRAPE aligns VLAs on a trajectory level and endows the model with the ability for global decision-making, instead of merely cloning behavior; 👉2. GRAPE implicitly models reward from both successful and failed trials to boost generalizability to diverse tasks; 👉3. GRAPE adopts a scalable preference synthesis algorithm to rank trajectories with preferences that align with arbitrary objectives. Our experiments on a diverse array of real-world and simulated robotic tasks reveal: 1⃣GRAPE enhances the performance of state-of-the-art VLA models, increasing success rates on in-domain and unseen manipulation tasks by 51.79% and 60.36%; 2⃣GRAPE is versatile to be aligned with diverse objectives and reduce collision rates by 44.31% or rollout length by 11.15% when aligning towards safer or more efficient manipulation policy, respectively. Check out our full project for more details: 🔥 Paper: 🔥 Project: 🔥 Code:

Huaxiu Yao

20,071 views • 1 year ago

AI TENNIS ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, PYTORCH, AND KEYPOINT EXTRACTION. Take any tennis match broadcast, any camera angle, any resolution. Feed it into the pipeline. YOLO detects both players and the tennis ball frame by frame. No manual labeling, no pre-annotated dataset. A fine-tuned YOLOv5 model trained on a Roboflow tennis ball dataset handles the ball - the hardest object to track in any sport. Tiny, fast, constantly occluded. The model finds it anyway. Trackers maintain identity across frames so Player 1 stays Player 1 from the first serve to match point. But detection is just the start. A ResNet50 CNN trained in PyTorch predicts court keypoints from every frame - the corners, service lines, baselines, net posts. Fourteen points that define the entire playing surface geometry. From those keypoints the system builds a homography matrix and warps the broadcast perspective into a top-down mini court with real coordinates. Now every player has a position in real space, not pixel space. Every frame becomes a measurement. Every rally becomes a dataset. Player movement speed - calculated from position deltas between frames, converted to meters per second through the homography. Ball shot speed - measured from the ball trajectory across consecutive detections. Number of shots per rally - counted automatically through ball direction changes. All of this rendered live on the video as an overlay. A mini court in the corner showing both players as dots moving in real time. Stats updating after every point. OpenCV handles the rendering. Pandas handles the math. PyTorch handles the intelligence. YOLO handles the eyes. No Hawkeye subscription, no court-embedded sensors, no tracking chips in the ball. A Python script, a trained model, and a GPU. The full code is on GitHub. The tutorial walks through every module - from ball detector training to court keypoint extraction to the final statistical overlay. Professional teams used to need broadcast deals and proprietary hardware for this kind of analysis. Now you build it in an afternoon with open-source tools. Trading here: Computer vision didn't just enter tennis. It made the expensive stuff free.

AI TENNIS ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, PYTORCH, AND KEYPOINT EXTRACTION. Take any tennis match broadcast, any camera angle, any resolution. Feed it into the pipeline. YOLO detects both players and the tennis ball frame by frame. No manual labeling, no pre-annotated dataset. A fine-tuned YOLOv5 model trained on a Roboflow tennis ball dataset handles the ball - the hardest object to track in any sport. Tiny, fast, constantly occluded. The model finds it anyway. Trackers maintain identity across frames so Player 1 stays Player 1 from the first serve to match point. But detection is just the start. A ResNet50 CNN trained in PyTorch predicts court keypoints from every frame - the corners, service lines, baselines, net posts. Fourteen points that define the entire playing surface geometry. From those keypoints the system builds a homography matrix and warps the broadcast perspective into a top-down mini court with real coordinates. Now every player has a position in real space, not pixel space. Every frame becomes a measurement. Every rally becomes a dataset. Player movement speed - calculated from position deltas between frames, converted to meters per second through the homography. Ball shot speed - measured from the ball trajectory across consecutive detections. Number of shots per rally - counted automatically through ball direction changes. All of this rendered live on the video as an overlay. A mini court in the corner showing both players as dots moving in real time. Stats updating after every point. OpenCV handles the rendering. Pandas handles the math. PyTorch handles the intelligence. YOLO handles the eyes. No Hawkeye subscription, no court-embedded sensors, no tracking chips in the ball. A Python script, a trained model, and a GPU. The full code is on GitHub. The tutorial walks through every module - from ball detector training to court keypoint extraction to the final statistical overlay. Professional teams used to need broadcast deals and proprietary hardware for this kind of analysis. Now you build it in an afternoon with open-source tools. Trading here: Computer vision didn't just enter tennis. It made the expensive stuff free.

zostaff

121,478 views • 5 months ago

I take back my statement that “Anyone can build a game with AI.” After 256 prompts building Top Tennis, I think people underestimate the person giving the instructions. One example from a prompt I wrote for Top Tennis: “i feel like the only way to score is going under the net or waiting for the CPU to make a mistake. there should be other ways. a deep ball should push the opponent behind the baseline, leaving space for a drop shot. after a difficult return, it should take longer to recover. changing direction should cost it time. give the CPU a reaction delay before it responds to my shot. it shouldn’t know where i’m aiming before racket contact. when it changes direction, make it decelerate before accelerating the other way. make return quality depend on positioning, incoming ball speed and preparation time. a stretched player running at full speed shouldn’t hit with the same accuracy and power as one waiting in position. look at player speed, acceleration, court size and ball speed together. compare how long the ball takes to reach its destination with how long the opponent needs to get there. test specific patterns: deep ball then drop shot, wide crosscourt then down the line, and hitting behind a moving opponent. check whether they create real openings without making every combination an automatic winner. beating the CPU should feel like you’re learning how to play tennis. analyze this deeply before changing anything.”

I take back my statement that “Anyone can build a game with AI.” After 256 prompts building Top Tennis, I think people underestimate the person giving the instructions. One example from a prompt I wrote for Top Tennis: “i feel like the only way to score is going under the net or waiting for the CPU to make a mistake. there should be other ways. a deep ball should push the opponent behind the baseline, leaving space for a drop shot. after a difficult return, it should take longer to recover. changing direction should cost it time. give the CPU a reaction delay before it responds to my shot. it shouldn’t know where i’m aiming before racket contact. when it changes direction, make it decelerate before accelerating the other way. make return quality depend on positioning, incoming ball speed and preparation time. a stretched player running at full speed shouldn’t hit with the same accuracy and power as one waiting in position. look at player speed, acceleration, court size and ball speed together. compare how long the ball takes to reach its destination with how long the opponent needs to get there. test specific patterns: deep ball then drop shot, wide crosscourt then down the line, and hitting behind a moving opponent. check whether they create real openings without making every combination an automatic winner. beating the CPU should feel like you’re learning how to play tennis. analyze this deeply before changing anything.”

Matteo V.

71,310 views • 16 days ago

China just automated HANDWRITING. Biden’s famous autopen now looks like a CHILDREN’S TOY next to this machine. This one is working in China TODAY!!, copying students’ handwritten notes with a real pen - line by line, symbol by symbol. But here’s the part that changes everything: It can reproduce Chinese characters, math formulas, diagrams, and entire pages of notes directly on paper. No printer. No copy-paste. A real pen physically writing the assignment. Give the machine a sample, feed it the content, and it can reproduce page after page while the student does something else. A few years ago, AI learned to write the answer. Now robots are learning to physically write it down for you. The line between “I did my homework” and “my machine did my homework” is about to disappear.

China just automated HANDWRITING. Biden’s famous autopen now looks like a CHILDREN’S TOY next to this machine. This one is working in China TODAY!!, copying students’ handwritten notes with a real pen - line by line, symbol by symbol. But here’s the part that changes everything: It can reproduce Chinese characters, math formulas, diagrams, and entire pages of notes directly on paper. No printer. No copy-paste. A real pen physically writing the assignment. Give the machine a sample, feed it the content, and it can reproduce page after page while the student does something else. A few years ago, AI learned to write the answer. Now robots are learning to physically write it down for you. The line between “I did my homework” and “my machine did my homework” is about to disappear.

Mutant CYT

34,767 views • 2 months ago

Tennis players live 9.7 years longer than sedentary people. Not 9.7 months. 9.7 years. Nearly a decade. The Copenhagen City Heart Study tracked 8,577 people for 25 years and ranked every sport by how much life it adds. Badminton: 6.2 years. Soccer: 4.7. Cycling: 3.7. Swimming: 3.4. Jogging: 3.2. Tennis almost triples jogging. A separate study of 80,000 adults found racket sports cut all-cause mortality by 47% and cardiovascular death by 56%. Swimming hit 41%. Aerobics hit 36%. The question is why racket sports destroy everything else. Three mechanisms stack on top of each other. First, the physical demands. A tennis rally requires explosive sprints, lateral cuts, and sustained aerobic output. You're training fast-twitch and slow-twitch muscle fibers simultaneously. Most cardio only trains one system. Second, the cognitive load. You're reading spin, predicting angles, adjusting position, and executing motor patterns in real-time. Your brain is solving spatial puzzles at 80+ mph. That hand-eye coordination and strategic processing builds neural connections that protect against cognitive decline. Third, and this is the one researchers keep coming back to: you literally cannot play alone. Every racket sport requires another person on the other side of the net. That forced social interaction triggers neurochemical benefits that solitary exercise cannot replicate. Strong social connection alone increases your chance of longevity by 50%. Jogging is you and your thoughts. Tennis is you, a strategic opponent, and a community. Dr. Daniel Amen is right. The data is overwhelming. If you want the single highest-ROI activity for a longer life, pick up a racket.

Tennis players live 9.7 years longer than sedentary people. Not 9.7 months. 9.7 years. Nearly a decade. The Copenhagen City Heart Study tracked 8,577 people for 25 years and ranked every sport by how much life it adds. Badminton: 6.2 years. Soccer: 4.7. Cycling: 3.7. Swimming: 3.4. Jogging: 3.2. Tennis almost triples jogging. A separate study of 80,000 adults found racket sports cut all-cause mortality by 47% and cardiovascular death by 56%. Swimming hit 41%. Aerobics hit 36%. The question is why racket sports destroy everything else. Three mechanisms stack on top of each other. First, the physical demands. A tennis rally requires explosive sprints, lateral cuts, and sustained aerobic output. You're training fast-twitch and slow-twitch muscle fibers simultaneously. Most cardio only trains one system. Second, the cognitive load. You're reading spin, predicting angles, adjusting position, and executing motor patterns in real-time. Your brain is solving spatial puzzles at 80+ mph. That hand-eye coordination and strategic processing builds neural connections that protect against cognitive decline. Third, and this is the one researchers keep coming back to: you literally cannot play alone. Every racket sport requires another person on the other side of the net. That forced social interaction triggers neurochemical benefits that solitary exercise cannot replicate. Strong social connection alone increases your chance of longevity by 50%. Jogging is you and your thoughts. Tennis is you, a strategic opponent, and a community. Dr. Daniel Amen is right. The data is overwhelming. If you want the single highest-ROI activity for a longer life, pick up a racket.

Aakash Gupta

5,245,831 views • 6 months ago

Medvedev was asked what it would take for players ranked No. 3 to No. 15 to challenge Sinner & Alcaraz, ‘Everyone can beat them on a given day… I’m not shy to say that if I play 10 matches each with Carlos & Jannik, I’ll probably lose a lot of them’ “I wanted to ask if the players ranked 3 to 15 are playing about the same level and can win on any day? And what it would take for that bunch to challenge Jannik and Carlos.” Daniil: “Jannik and Carlos it’s tough. They just play better tennis than other guys. I think what’s important is that I played a lot of big names in tennis, and you can beat anyone on a given day. I’m not shy to say that if I play 20 matches against Carlos and Jannik, let’s say 10 with each of them, I’ll probably lose a lot of them. But I’m gonna try my best in every one of them to win. Out of 10 matches, you can win some. I’m not gonna say a number, but you can win. They can have a bit of an off day. I did beat both of them in Grand Slams.” (via Australian Open Press)

Medvedev was asked what it would take for players ranked No. 3 to No. 15 to challenge Sinner & Alcaraz, ‘Everyone can beat them on a given day… I’m not shy to say that if I play 10 matches each with Carlos & Jannik, I’ll probably lose a lot of them’ “I wanted to ask if the players ranked 3 to 15 are playing about the same level and can win on any day? And what it would take for that bunch to challenge Jannik and Carlos.” Daniil: “Jannik and Carlos it’s tough. They just play better tennis than other guys. I think what’s important is that I played a lot of big names in tennis, and you can beat anyone on a given day. I’m not shy to say that if I play 20 matches against Carlos and Jannik, let’s say 10 with each of them, I’ll probably lose a lot of them. But I’m gonna try my best in every one of them to win. Out of 10 matches, you can win some. I’m not gonna say a number, but you can win. They can have a bit of an off day. I did beat both of them in Grand Slams.” (via Australian Open Press)

The Tennis Letter

600,709 views • 8 months ago

Alignment trouble looks like a setup problem. Often it's a perception problem. A right-handed, right-eye dominant player often has more difficulty aligning to the target than a cross-dominant player. If alignment is an issue, here are two of our favorite strategies that Dr. Rose introduces in our advanced Golf curriculum. 1) Move the reference spot behind the ball: Instead of picking a reference spot in front of the ball, pick a spot behind the ball on the target line and walk in looking at that. For same-side dominant players, this moves the spot closer to their dominant eye. 2) Keep it binocular: Pick a spot in front and keep your eye on it as you walk in. Draw the line from the ball through the spot, square yourself to it, then walk into the shot. This keeps both eyes working together the whole time. Poor alignment can quietly hinder progress, and a player's visual system can make it even harder to correct. Our next Golf Level 3 seminar is October 9th - 11th at TPI HQ. Golf Level 2 is available via online course.

Alignment trouble looks like a setup problem. Often it's a perception problem. A right-handed, right-eye dominant player often has more difficulty aligning to the target than a cross-dominant player. If alignment is an issue, here are two of our favorite strategies that Dr. Rose introduces in our advanced Golf curriculum. 1) Move the reference spot behind the ball: Instead of picking a reference spot in front of the ball, pick a spot behind the ball on the target line and walk in looking at that. For same-side dominant players, this moves the spot closer to their dominant eye. 2) Keep it binocular: Pick a spot in front and keep your eye on it as you walk in. Draw the line from the ball through the spot, square yourself to it, then walk into the shot. This keeps both eyes working together the whole time. Poor alignment can quietly hinder progress, and a player's visual system can make it even harder to correct. Our next Golf Level 3 seminar is October 9th - 11th at TPI HQ. Golf Level 2 is available via online course.

TPI

50,417 views • 3 months ago

🚨 BREAKING: Microsoft's first robotics foundation model! 🤯 Microsoft just announced Rho-alpha (ρα), their first robotics model derived from the Phi series of vision-language models. Rho-alpha translates natural language commands into control signals for robotic systems performing bimanual manipulation tasks. Commands like "push the green button with the right gripper," "pull out the red wire," "flip the top switch on," or "turn the knob to position 5" get executed directly by dual-arm robots. What makes this different from standard vision-language-action (VLA) models is the additional modalities. Rho-alpha is a VLA+ model that adds tactile sensing to the perceptual mix, with plans to incorporate force feedback. On the learning side, the model is designed to continually improve during deployment by learning from human feedback. The training approach combines trajectories from physical demonstrations and simulated tasks with web-scale visual question answering data. Since teleoperation data is scarce and expensive, Microsoft is using NVIDIA Isaac Sim on Azure to generate physically accurate synthetic datasets via reinforcement learning. These simulated trajectories get combined with commercial and open physical demonstration datasets. The model is currently under evaluation on dual-arm setups and humanoid robots. Microsoft is opening an Early Access Program for organizations interested in evaluating Rho-alpha. Robots that can adapt to dynamic situations and human preferences are more useful in real environments and more trusted by the people operating them. Read more here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

🚨 BREAKING: Microsoft's first robotics foundation model! 🤯 Microsoft just announced Rho-alpha (ρα), their first robotics model derived from the Phi series of vision-language models. Rho-alpha translates natural language commands into control signals for robotic systems performing bimanual manipulation tasks. Commands like "push the green button with the right gripper," "pull out the red wire," "flip the top switch on," or "turn the knob to position 5" get executed directly by dual-arm robots. What makes this different from standard vision-language-action (VLA) models is the additional modalities. Rho-alpha is a VLA+ model that adds tactile sensing to the perceptual mix, with plans to incorporate force feedback. On the learning side, the model is designed to continually improve during deployment by learning from human feedback. The training approach combines trajectories from physical demonstrations and simulated tasks with web-scale visual question answering data. Since teleoperation data is scarce and expensive, Microsoft is using NVIDIA Isaac Sim on Azure to generate physically accurate synthetic datasets via reinforcement learning. These simulated trajectories get combined with commercial and open physical demonstration datasets. The model is currently under evaluation on dual-arm setups and humanoid robots. Microsoft is opening an Early Access Program for organizations interested in evaluating Rho-alpha. Robots that can adapt to dynamic situations and human preferences are more useful in real environments and more trusted by the people operating them. Read more here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

61,071 views • 8 months ago

In one of the coolest moments in rock history, Eddie Van Halen stood inside the Smithsonian’s National Museum of American History and broke down exactly how he invented his legendary finger-tapping technique. He explained how he stumbled on the idea as a kid — inspired by piano and classical techniques — and later saw Jimmy Page playing with one hand and turned it into the explosive sound that changed guitar forever. Then he plugged in and unleashed “Eruption” live, right there in the museum. Watching Eddie demonstrate the birth of modern tapping in such a historic setting? Pure magic. That two-handed technique went from his bedroom to stadiums worldwide and influenced an entire generation of players. Eddie didn’t just play guitar — he rewrote the rules. One of the greatest innovators rock has ever seen. Rest in power, EVH.

In one of the coolest moments in rock history, Eddie Van Halen stood inside the Smithsonian’s National Museum of American History and broke down exactly how he invented his legendary finger-tapping technique. He explained how he stumbled on the idea as a kid — inspired by piano and classical techniques — and later saw Jimmy Page playing with one hand and turned it into the explosive sound that changed guitar forever. Then he plugged in and unleashed “Eruption” live, right there in the museum. Watching Eddie demonstrate the birth of modern tapping in such a historic setting? Pure magic. That two-handed technique went from his bedroom to stadiums worldwide and influenced an entire generation of players. Eddie didn’t just play guitar — he rewrote the rules. One of the greatest innovators rock has ever seen. Rest in power, EVH.

Roy Rogers Happy Trails Music Shop

363,309 views • 3 months ago

Mojo Australian Open Quarterfinals 🧙🏼‍♂️🇦🇺🎾 —— 5U MAX 2026: (18-3) 85.71% 🏦 —— 5U 🇺🇸Ben Shelton +7.0 games (-155) 🔮 Seven games for Shelton when he is arguably playing the best tennis of his career is too many. Shelton has only been broken 3 times this entire tournament and has played better overall players by a landslide than Sinner. Shelton has hit 184 winners to 115 unforced errors in his 4 matches in Melbourne. This stat alone shows the quality of tennis Shelton has been playing. Sinner has hit 152 winners to 99 unforced errors. Shelton has a slight advantage in winners/unforced errors ratio while playing better opponents… And we are getting 7 games. Look, I know Sinner is insane and I know the history between these two players. I will tell you I have watched Shelton play shout 100 matches of professional tennis and the confidence he is playing with right now is unlike anything I have seen from him. While on the contrary Sinner has not looked anything above his typical level. Give me the 7 games. Even if Shelton gets broken once in every single set, we still cover or push. 6-3 6-4 6-4 | 6-4 6-4 6-4 . And this is assume we don’t get any break chances of our own. If Shelton can confidently hold serve like he has been doing all tournament then it will open the door for him to be hyper aggressive on return and go for shots that will win him the match. I love this play. Let’s get it 🧙🏼‍♂️

Mojo Australian Open Quarterfinals 🧙🏼‍♂️🇦🇺🎾 —— 5U MAX 2026: (18-3) 85.71% 🏦 —— 5U 🇺🇸Ben Shelton +7.0 games (-155) 🔮 Seven games for Shelton when he is arguably playing the best tennis of his career is too many. Shelton has only been broken 3 times this entire tournament and has played better overall players by a landslide than Sinner. Shelton has hit 184 winners to 115 unforced errors in his 4 matches in Melbourne. This stat alone shows the quality of tennis Shelton has been playing. Sinner has hit 152 winners to 99 unforced errors. Shelton has a slight advantage in winners/unforced errors ratio while playing better opponents… And we are getting 7 games. Look, I know Sinner is insane and I know the history between these two players. I will tell you I have watched Shelton play shout 100 matches of professional tennis and the confidence he is playing with right now is unlike anything I have seen from him. While on the contrary Sinner has not looked anything above his typical level. Give me the 7 games. Even if Shelton gets broken once in every single set, we still cover or push. 6-3 6-4 6-4 | 6-4 6-4 6-4 . And this is assume we don’t get any break chances of our own. If Shelton can confidently hold serve like he has been doing all tournament then it will open the door for him to be hyper aggressive on return and go for shots that will win him the match. I love this play. Let’s get it 🧙🏼‍♂️

MOJO 🧙🏼‍♂️

37,303 views • 8 months ago