Загрузка видео...

Не удалось загрузить видео

На главную

Weights drop ⚠️ We released our pre-trained model for the cup arrangement task trained on 1400 demos! We aim to enable anyone to deploy UMI on their robot to arrange any "espresso cup with saucer" they buy on Amazon.

chichengcc's profile picture

Cheng Chi

9,859 subscribers

25,137 просмотров • 2 лет назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Many of you asked for code & weights for π₀, we are happy to announce that we are releasing π₀ and pre-trained checkpoints in our new openpi repository! We tested the model on a few public robots, and we include code for you to fine-tune it yourself.
0:20

Many of you asked for code & weights for π₀, we are happy to announce that we are releasing π₀ and pre-trained checkpoints in our new openpi repository! We tested the model on a few public robots, and we include code for you to fine-tune it yourself.

Physical Intelligence

441,382 просмотров • 1 год назад

🤖What if a robot could perform a new task just from a natural language command, with zero demonstrations? Our new work, NovaFlow, makes it possible! We use pre-trained video generative model to create a video of the task, then translate it into a plan for real-world robot execution. 1/6 #Robotics #AI #ZeroShot #Manipulation
0:27

🤖What if a robot could perform a new task just from a natural language command, with zero demonstrations? Our new work, NovaFlow, makes it possible! We use pre-trained video generative model to create a video of the task, then translate it into a plan for real-world robot execution. 1/6 #Robotics #AI #ZeroShot #Manipulation

Hongyu Li

105,669 просмотров • 10 месяцев назад

#WATCH | Delhi | Bhairav Battalion Contingent Commander says, "This is for the first time that a contingent is marching at the Republic Day parade and Army Day parade within 1 year of its formation. We feel proud and we are ready for this opportunity... We are trained as per the task we will be provided... We are trained to complete any task on any terrain..."
1:10

#WATCH | Delhi | Bhairav Battalion Contingent Commander says, "This is for the first time that a contingent is marching at the Republic Day parade and Army Day parade within 1 year of its formation. We feel proud and we are ready for this opportunity... We are trained as per the task we will be provided... We are trained to complete any task on any terrain..."

ANI

139,636 просмотров • 6 месяцев назад

We just released Firefly Custom Models (beta) to everyone 🎉 You can train a model on your own images to keep a consistent style or character! I trained a model on my photography and generated images that would be impossible to get any other way for my upcoming film.
0:32

We just released Firefly Custom Models (beta) to everyone 🎉 You can train a model on your own images to keep a consistent style or character! I trained a model on my photography and generated images that would be impossible to get any other way for my upcoming film.

Kris Kashtanova

26,766 просмотров • 4 месяцев назад

We're releasing Paris 2.0, which, to our knowledge, is the world's first decentralized trained video generation model. We benchmarked it against a monolithic model trained on the same data and compute budget, and Paris 2.0 outperformed the monolithic by ~2x on FVD benchmark.
0:42

We're releasing Paris 2.0, which, to our knowledge, is the world's first decentralized trained video generation model. We benchmarked it against a monolithic model trained on the same data and compute budget, and Paris 2.0 outperformed the monolithic by ~2x on FVD benchmark.

Bidhan Roy

456,522 просмотров • 2 месяцев назад

VLAs offer an avenue for generalist robot policies; however, naively following the action predictions leads to brittle or unsafe behaviours. We introduce VLAPS, which integrates model-based search with pre-trained VLA policies to improve performance without additional training.
0:30

VLAs offer an avenue for generalist robot policies; however, naively following the action predictions leads to brittle or unsafe behaviours. We introduce VLAPS, which integrates model-based search with pre-trained VLA policies to improve performance without additional training.

Glen Berseth

13,374 просмотров • 11 месяцев назад

Excited to release τ0-WM: an open-source unified video-action world model for robotic manipulation. It's a 5B-parameter robotic foundation model trained on 27.3K hours of real-robot teleoperation, UMI-style demonstrations, and egocentric interaction videos.
3:26

Excited to release τ0-WM: an open-source unified video-action world model for robotic manipulation. It's a 5B-parameter robotic foundation model trained on 27.3K hours of real-robot teleoperation, UMI-style demonstrations, and egocentric interaction videos.

Jianlan Luo

63,831 просмотров • 2 месяцев назад

Our strategic partnership with the #EsportsWorldCup underscores our dedication to uniting people through their shared passion for gaming We invest in people’s potential today to enable a brighter tomorrow 🎮 Esports World Cup #aramco
0:27

Our strategic partnership with the #EsportsWorldCup underscores our dedication to uniting people through their shared passion for gaming We invest in people’s potential today to enable a brighter tomorrow 🎮 Esports World Cup #aramco

aramco

7,890,062 просмотров • 2 лет назад

Sharing a few more demos of our first #LargeBehaviorModel (LBM) at TRI. 1/🍎This model enables a robot to learned to core and cut an apple into multiple slices autonomously. We trained our diffusion-based LBM on almost 1,700 hours of robot data, conducted 1,800 real-world evaluation rollouts, and ran over 47,000 simulation rollouts to rigorously study its capabilities. Learn more here:
2:04

Sharing a few more demos of our first #LargeBehaviorModel (LBM) at TRI. 1/🍎This model enables a robot to learned to core and cut an apple into multiple slices autonomously. We trained our diffusion-based LBM on almost 1,700 hours of robot data, conducted 1,800 real-world evaluation rollouts, and ran over 47,000 simulation rollouts to rigorously study its capabilities. Learn more here:

Zubair Irshad

12,202 просмотров • 1 год назад

JetBlue Flight 1112 from Curaçao to JFK almost collided with a U.S. Air Force refueling tanker on Friday, with the pilot heard saying to ATC “We almost had a midair collision up here, they passed directly in our flight path. … They don’t have their transponder turned on, it’s outrageous.” Derek Dombrowski, a spokesman for JetBlue, said Sunday: “We have reported this incident to federal authorities and will participate in any investigation." He added, “Our crewmembers are trained on proper procedures for various flight situations, and we appreciate our crew for promptly reporting this situation to our leadership team.” SOUTHCOM released the following statement: “We are aware of the recent reporting regarding U.S. military aircraft operations in the Caribbean and are currently reviewing the matter. Military aircrews are highly trained professionals who operate in accordance with established procedures and applicable airspace requirements. Safety remains a top priority, and we are working through the appropriate channels to assess the facts surrounding the situation.”
2:00

JetBlue Flight 1112 from Curaçao to JFK almost collided with a U.S. Air Force refueling tanker on Friday, with the pilot heard saying to ATC “We almost had a midair collision up here, they passed directly in our flight path. … They don’t have their transponder turned on, it’s outrageous.” Derek Dombrowski, a spokesman for JetBlue, said Sunday: “We have reported this incident to federal authorities and will participate in any investigation." He added, “Our crewmembers are trained on proper procedures for various flight situations, and we appreciate our crew for promptly reporting this situation to our leadership team.” SOUTHCOM released the following statement: “We are aware of the recent reporting regarding U.S. military aircraft operations in the Caribbean and are currently reviewing the matter. Military aircrews are highly trained professionals who operate in accordance with established procedures and applicable airspace requirements. Safety remains a top priority, and we are working through the appropriate channels to assess the facts surrounding the situation.”

Breaking Aviation News & Videos

272,514 просмотров • 8 месяцев назад

“Very few trainers from Europe have trained Breeders' Cup winners,” said trainer Donnacha O'Brien. “I think Dad is the only Irish trainer that has trained more than one Breeders' Cup winner, so to be on that list is very special and it shows you how hard it is to do. It's something that we'll always have on the CV and hopefully we can add a few more.” Hear from Donnacha on his Breeders' Cup Breakthrough at Del Mar with Balantina in the Juvenile Fillies Turf Produced with TDN
3:09

“Very few trainers from Europe have trained Breeders' Cup winners,” said trainer Donnacha O'Brien. “I think Dad is the only Irish trainer that has trained more than one Breeders' Cup winner, so to be on that list is very special and it shows you how hard it is to do. It's something that we'll always have on the CV and hopefully we can add a few more.” Hear from Donnacha on his Breeders' Cup Breakthrough at Del Mar with Balantina in the Juvenile Fillies Turf Produced with TDN

Breeders' Cup

11,043 просмотров • 2 месяцев назад

A humanoid robot policy trained solely on synthetic data generated by a world model. Research Scientist Joel Jang presents NVIDIA's DreamGen pipeline: ⦿ Post-train the world model Cosmos-Predict2 with a small set of real teleoperation demos. ⦿ Prompt the world model to generate synthetic video data with verbs and scenarios not used in the world model’s post-training. ⦿ Auto-label synthetic video data with action sequences. ⦿ Train robot policies using only synthetic data. That's it. Deploy zero-shot to a real humanoid robot.
2:05

A humanoid robot policy trained solely on synthetic data generated by a world model. Research Scientist Joel Jang presents NVIDIA's DreamGen pipeline: ⦿ Post-train the world model Cosmos-Predict2 with a small set of real teleoperation demos. ⦿ Prompt the world model to generate synthetic video data with verbs and scenarios not used in the world model’s post-training. ⦿ Auto-label synthetic video data with action sequences. ⦿ Train robot policies using only synthetic data. That's it. Deploy zero-shot to a real humanoid robot.

The Humanoid Hub

20,968 просмотров • 1 год назад

AppleScript is the oldest way to drive a Mac. Frontier models butcher it. So we trained two small ones that don't. Open weights. On-device. 100% compile.
0:20

AppleScript is the oldest way to drive a Mac. Frontier models butcher it. So we trained two small ones that don't. Open weights. On-device. 100% compile.

Osaurus

52,628 просмотров • 1 месяц назад

Rep Her just said on the House Floor that we shouldn't allow women to keep guns in their own homes for self-defense because they aren't trained well enough.
2:19

Rep Her just said on the House Floor that we shouldn't allow women to keep guns in their own homes for self-defense because they aren't trained well enough.

MN House GOP Rapid Response

196,796 просмотров • 2 лет назад

These guys trained a robot with water bottles, and the robot learned to pick up other objects it had never practiced on. They did this using reinforcement learning: 1. The robot had to pick a bottle out of a messy tote 2. Then it had to hand the bottle to a person 3. Only water bottles. Nothing else. After they finished training, they tested the robot on different objects: • On bottles, the success went from 80% to 98% • On snack cans, the robot's throughput rose 40% • On soft bags, the robot's performance improved by 13% even though they're deformable and much harder to handle than rigid objects. Snack cans and soft bags have different shapes and weights, and the robot was never trained on them. This was just one of three use cases they tested. Every hour we spend teaching a robot costs a ton of money. If they only get better at handling the one object they practice on, reinforcement learning will never be worth doing. But if the gains carry over, we could focus on teaching the hardest possible tasks and let the robot apply that knowledge elsewhere. Here is the full write-up by Humanoid on the infrastructure they used and what they learned: #ad
1:40

These guys trained a robot with water bottles, and the robot learned to pick up other objects it had never practiced on. They did this using reinforcement learning: 1. The robot had to pick a bottle out of a messy tote 2. Then it had to hand the bottle to a person 3. Only water bottles. Nothing else. After they finished training, they tested the robot on different objects: • On bottles, the success went from 80% to 98% • On snack cans, the robot's throughput rose 40% • On soft bags, the robot's performance improved by 13% even though they're deformable and much harder to handle than rigid objects. Snack cans and soft bags have different shapes and weights, and the robot was never trained on them. This was just one of three use cases they tested. Every hour we spend teaching a robot costs a ton of money. If they only get better at handling the one object they practice on, reinforcement learning will never be worth doing. But if the gains carry over, we could focus on teaching the hardest possible tasks and let the robot apply that knowledge elsewhere. Here is the full write-up by Humanoid on the infrastructure they used and what they learned: #ad

Santiago

13,952 просмотров • 27 дней назад

"We'll do what we need to do to get through the group" Steven Naismith on what Scotland need to build on ahead of their World Cup clash with Brazil 🇧🇷
1:44

"We'll do what we need to do to get through the group" Steven Naismith on what Scotland need to build on ahead of their World Cup clash with Brazil 🇧🇷

Sky Sports Scotland

109,563 просмотров • 1 месяц назад

Sec. Mullin on plans to deploy National Guardsmen to assist with security at FIFA World Cup venues: “We're going to be flexing hard and using every asset that we have.”
1:48

Sec. Mullin on plans to deploy National Guardsmen to assist with security at FIFA World Cup venues: “We're going to be flexing hard and using every asset that we have.”

Breaking911

66,328 просмотров • 3 месяцев назад

Kangana Ranaut on cricketer Vaibhav Sooryavanshi : We really wish him the best. And we want him to be better than Sachin or Virat for our country. I mean not to put any pressure on him but bring the World Cup or something. #VaibhavSooryavanshi #KanganaRanaut
0:29

Kangana Ranaut on cricketer Vaibhav Sooryavanshi : We really wish him the best. And we want him to be better than Sachin or Virat for our country. I mean not to put any pressure on him but bring the World Cup or something. #VaibhavSooryavanshi #KanganaRanaut

Rahul Chauhan

75,301 просмотров • 2 месяцев назад