正在加载视频...

视频加载失败

返回首页

Everyone in Embodied AI is talking about Vision-Language-Action (VLA) models. Almost no one is talking about the physical nightmare of collecting the data to train them. You can't scrape a kitchen table or a warehouse shelf from a web browser. To get to millions of hours of diverse, real-world...

ryanmhickman's profile picture

Ryan Hickman

3,212 subscribers

26,910 次观看 • 3 个月前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Exciting updates on Project GR00T! We discover a systematic way to scale up robot data, tackling the most painful pain point in robotics. The idea is simple: human collects demonstration on a real robot, and we multiply that data 1000x or more in simulation. Let’s break it down: 1. We use Apple Vision Pro (yes!!) to give the human operator first person control of the humanoid. Vision Pro parses human hand pose and retargets the motion to the robot hand, all in real time. From the human’s point of view, they are immersed in another body like the Avatar. Teleoperation is slow and time-consuming, but we can afford to collect a small amount of data. 2. We use RoboCasa, a generative simulation framework, to multiply the demonstration data by varying the visual appearance and layout of the environment. In Jensen’s keynote video below, the humanoid is now placing the cup in hundreds of kitchens with a huge diversity of textures, furniture, and object placement. We only have 1 physical kitchen at the GEAR Lab in NVIDIA HQ, but we can conjure up infinite ones in simulation. 3. Finally, we apply MimicGen, a technique to multiply the above data even more by varying the *motion* of the robot. MimicGen generates vast number of new action trajectories based on the original human data, and filters out failed ones (e.g. those that drop the cup) to form a much larger dataset. To sum up, given 1 human trajectory with Vision Pro -> RoboCasa produces N (varying visuals) -> MimicGen further augments to NxM (varying motions). This is the way to trade compute for expensive human data by GPU-accelerated simulation. A while ago, I mentioned that teleoperation is fundamentally not scalable, because we are always limited by 24 hrs/robot/day in the world of atoms. Our new GR00T synthetic data pipeline breaks this barrier in the world of bits. Scaling has been so much fun for LLMs, and it's finally our turn to have fun in robotics! We are building tools to enable everyone in the ecosystem to scale up with us. Links in thread:
1:21

Exciting updates on Project GR00T! We discover a systematic way to scale up robot data, tackling the most painful pain point in robotics. The idea is simple: human collects demonstration on a real robot, and we multiply that data 1000x or more in simulation. Let’s break it down: 1. We use Apple Vision Pro (yes!!) to give the human operator first person control of the humanoid. Vision Pro parses human hand pose and retargets the motion to the robot hand, all in real time. From the human’s point of view, they are immersed in another body like the Avatar. Teleoperation is slow and time-consuming, but we can afford to collect a small amount of data. 2. We use RoboCasa, a generative simulation framework, to multiply the demonstration data by varying the visual appearance and layout of the environment. In Jensen’s keynote video below, the humanoid is now placing the cup in hundreds of kitchens with a huge diversity of textures, furniture, and object placement. We only have 1 physical kitchen at the GEAR Lab in NVIDIA HQ, but we can conjure up infinite ones in simulation. 3. Finally, we apply MimicGen, a technique to multiply the above data even more by varying the *motion* of the robot. MimicGen generates vast number of new action trajectories based on the original human data, and filters out failed ones (e.g. those that drop the cup) to form a much larger dataset. To sum up, given 1 human trajectory with Vision Pro -> RoboCasa produces N (varying visuals) -> MimicGen further augments to NxM (varying motions). This is the way to trade compute for expensive human data by GPU-accelerated simulation. A while ago, I mentioned that teleoperation is fundamentally not scalable, because we are always limited by 24 hrs/robot/day in the world of atoms. Our new GR00T synthetic data pipeline breaks this barrier in the world of bits. Scaling has been so much fun for LLMs, and it's finally our turn to have fun in robotics! We are building tools to enable everyone in the ecosystem to scale up with us. Links in thread:

Jim Fan

364,670 次观看 • 2 年前

We are publishing our first deep dive on what we believe is one of the most challenging layers in egocentric data - SLAM and VIO in the context of long-horizon state tracking. We break down how SLAM and VIO fail in egocentric settings - visual features vanish at close range, depth sensors saturate, fast head motion blurs frames, and these failures don't always occur in isolation. They hit at the exact same moment, leading to compounding errors and making the downstream data unusable. We believe the foundation for high-quality egocentric data demands sub-centimeter precision over long episodes ranging from a few minutes to up to an hour.
0:33

We are publishing our first deep dive on what we believe is one of the most challenging layers in egocentric data - SLAM and VIO in the context of long-horizon state tracking. We break down how SLAM and VIO fail in egocentric settings - visual features vanish at close range, depth sensors saturate, fast head motion blurs frames, and these failures don't always occur in isolation. They hit at the exact same moment, leading to compounding errors and making the downstream data unusable. We believe the foundation for high-quality egocentric data demands sub-centimeter precision over long episodes ranging from a few minutes to up to an hour.

FPV Labs

42,211 次观看 • 5 个月前

Physics-based Motion Retargeting from Sparse Inputs paper page: Avatars are important to create interactive and immersive experiences in virtual worlds. One challenge in animating these characters to mimic a user's motion is that commercial AR/VR products consist only of a headset and controllers, providing very limited sensor data of the user's pose. Another challenge is that an avatar might have a different skeleton structure than a human and the mapping between them is unclear. In this work we address both of these challenges. We introduce a method to retarget motions in real-time from sparse human sensor data to characters of various morphologies. Our method uses reinforcement learning to train a policy to control characters in a physics simulator. We only require human motion capture data for training, without relying on artist-generated animations for each avatar. This allows us to use large motion capture datasets to train general policies that can track unseen users from real and sparse data in real-time. We demonstrate the feasibility of our approach on three characters with different skeleton structure: a dinosaur, a mouse-like creature and a human. We show that the avatar poses often match the user surprisingly well, despite having no sensor information of the lower body available. We discuss and ablate the important components in our framework, specifically the kinematic retargeting step, the imitation, contact and action reward as well as our asymmetric actor-critic observations. We further explore the robustness of our method in a variety of settings including unbalancing, dancing and sports motions.
7:12

Physics-based Motion Retargeting from Sparse Inputs paper page: Avatars are important to create interactive and immersive experiences in virtual worlds. One challenge in animating these characters to mimic a user's motion is that commercial AR/VR products consist only of a headset and controllers, providing very limited sensor data of the user's pose. Another challenge is that an avatar might have a different skeleton structure than a human and the mapping between them is unclear. In this work we address both of these challenges. We introduce a method to retarget motions in real-time from sparse human sensor data to characters of various morphologies. Our method uses reinforcement learning to train a policy to control characters in a physics simulator. We only require human motion capture data for training, without relying on artist-generated animations for each avatar. This allows us to use large motion capture datasets to train general policies that can track unseen users from real and sparse data in real-time. We demonstrate the feasibility of our approach on three characters with different skeleton structure: a dinosaur, a mouse-like creature and a human. We show that the avatar poses often match the user surprisingly well, despite having no sensor information of the lower body available. We discuss and ablate the important components in our framework, specifically the kinematic retargeting step, the imitation, contact and action reward as well as our asymmetric actor-critic observations. We further explore the robustness of our method in a variety of settings including unbalancing, dancing and sports motions.

AK

106,527 次观看 • 3 年前

Gabe on why he doesn't think Anthropic or OpenAI will own finance: "The reason people get confused when they look at app layer businesses like mine or Harvey or Legora or Sierra is because there's a spectrum of perpendicularity to what the labs are building. There's a whole bunch of stuff underneath the surface that the labs are never going to build, that we need to build for finance. All of finance is a collection of different niches with different data sets, different regulatory requirements. And we can get to $5 billion in revenue by going deep across those things and creating the systems of record that help manage them. That for Anthropic would be like stopping on the side of the road to pick up a penny, because they're on the pathway of trying to go from $100 billion in revenue to a trillion in revenue. Say you are a big public company buying another big public company and you need to send data back and forth. You actually need some sort of data room, something that is compliant, safe and secure. And I don't think OpenAI or Anthropic will ever want to build a data room business. If you actually want to be the exchange for all of high finance, you don't just need to own the intelligence, you need to own the transaction venue, the communication venue, the workflows, and all the data inputs that go into it. Think about the fundamental difference between Claude Code when it came out versus ChatGPT. The models were actually fairly similar, but the harness and the way that it was presented from Claude was far better. The way that you harness these models is so, so important."
0:48

Gabe on why he doesn't think Anthropic or OpenAI will own finance: "The reason people get confused when they look at app layer businesses like mine or Harvey or Legora or Sierra is because there's a spectrum of perpendicularity to what the labs are building. There's a whole bunch of stuff underneath the surface that the labs are never going to build, that we need to build for finance. All of finance is a collection of different niches with different data sets, different regulatory requirements. And we can get to $5 billion in revenue by going deep across those things and creating the systems of record that help manage them. That for Anthropic would be like stopping on the side of the road to pick up a penny, because they're on the pathway of trying to go from $100 billion in revenue to a trillion in revenue. Say you are a big public company buying another big public company and you need to send data back and forth. You actually need some sort of data room, something that is compliant, safe and secure. And I don't think OpenAI or Anthropic will ever want to build a data room business. If you actually want to be the exchange for all of high finance, you don't just need to own the intelligence, you need to own the transaction venue, the communication venue, the workflows, and all the data inputs that go into it. Think about the fundamental difference between Claude Code when it came out versus ChatGPT. The models were actually fairly similar, but the harness and the way that it was presented from Claude was far better. The way that you harness these models is so, so important."

Patrick OShaughnessy

85,246 次观看 • 10 天前

The largest fully annotated open egocentric human dataset. Today we're open-sourcing EgoSuite-Open100K with Hugging Face : 100,000 hours of egocentric human data, free for anyone to use. Physical AI has a scaling law of its own now, and its input is human data. That part is settled. What the field still needs is a shared foundation to build on. 15,000+ tasks. 15,000+ real scenes, from factory floors to retail backrooms. Hand pose, body pose and subtask-level semantics, with wrist camera coverage on part of the set. Licensed for commercial training, not research-only. We build the data infrastructure for Physical AI. Open100K is the first public layer of it. The first 10,000 hours are live now, with the rest rolling out in stages.
1:18

The largest fully annotated open egocentric human dataset. Today we're open-sourcing EgoSuite-Open100K with Hugging Face : 100,000 hours of egocentric human data, free for anyone to use. Physical AI has a scaling law of its own now, and its input is human data. That part is settled. What the field still needs is a shared foundation to build on. 15,000+ tasks. 15,000+ real scenes, from factory floors to retail backrooms. Hand pose, body pose and subtask-level semantics, with wrist camera coverage on part of the set. Licensed for commercial training, not research-only. We build the data infrastructure for Physical AI. Open100K is the first public layer of it. The first 10,000 hours are live now, with the rest rolling out in stages.

Lightwheel

74,327 次观看 • 1 个月前

We’re on the verge of interactive, real-time, photorealistic video generation with what are called World Models. These require a fair amount of expensive compute, but costs will come down over time. Today, these interactions are essentially real-time dreams. They lack the persistence and shared logic that turns a video into a multiplayer game. Or concert. Or classroom. Or the holodeck. We believe the ultimate architecture for gaming (and the holodeck) is Roblox Reality. It’s a hybrid architecture that marries the structured data and logic of the Roblox Engine and Roblox Cloud with the generative power of Video World Models. The Roblox Engine provides the underlying synchronized ground truth—the score, physics, multiplayer sync, etc.—while our video model acts as a Super Upsampler to layer on photorealistic detail. We believe this will ultimately remove barriers to high-fidelity creation, allowing a team of three people to build a narrative-driven, photorealistic masterpiece in a single week. This is an early look at turning solitary AI dreams into a social, playable reality.
0:31

We’re on the verge of interactive, real-time, photorealistic video generation with what are called World Models. These require a fair amount of expensive compute, but costs will come down over time. Today, these interactions are essentially real-time dreams. They lack the persistence and shared logic that turns a video into a multiplayer game. Or concert. Or classroom. Or the holodeck. We believe the ultimate architecture for gaming (and the holodeck) is Roblox Reality. It’s a hybrid architecture that marries the structured data and logic of the Roblox Engine and Roblox Cloud with the generative power of Video World Models. The Roblox Engine provides the underlying synchronized ground truth—the score, physics, multiplayer sync, etc.—while our video model acts as a Super Upsampler to layer on photorealistic detail. We believe this will ultimately remove barriers to high-fidelity creation, allowing a team of three people to build a narrative-driven, photorealistic masterpiece in a single week. This is an early look at turning solitary AI dreams into a social, playable reality.

David Baszucki

790,990 次观看 • 5 个月前

The biggest marketplace on Earth won't be the assets humans own. It'll be the data AI needs. Everything that can be digitised is getting pulled in to train the next models. Every scientific dataset, every piece of research sitting in a university, every sensor reading off every John Deere tractor in a field. All of it feeds the machine, and almost none of it is visible to us. What comes out is an agentic economy running at machine speed. You plug your data into a vault, and agents monetise it on your behalf, trading with other agents in real time, in a marketplace we never see. The company sitting on proprietary data becomes part of it without even knowing. This is what the Economic Singularity actually looks like from the inside. The real economy stops being the one we can see and price, and moves somewhere we can't follow. You own a piece of it, or you get priced out of it. Jordi Visser.
1:27

The biggest marketplace on Earth won't be the assets humans own. It'll be the data AI needs. Everything that can be digitised is getting pulled in to train the next models. Every scientific dataset, every piece of research sitting in a university, every sensor reading off every John Deere tractor in a field. All of it feeds the machine, and almost none of it is visible to us. What comes out is an agentic economy running at machine speed. You plug your data into a vault, and agents monetise it on your behalf, trading with other agents in real time, in a marketplace we never see. The company sitting on proprietary data becomes part of it without even knowing. This is what the Economic Singularity actually looks like from the inside. The real economy stops being the one we can see and price, and moves somewhere we can't follow. You own a piece of it, or you get priced out of it. Jordi Visser.

Raoul Pal

79,047 次观看 • 2 个月前

#BudgetSession2025 | Lok Sabha LoP and Congress MP Rahul Gandhi says "People talk about AI, but it's important to understand that AI on its own is absolutely meaningless because AI operates on top of data. Without data, AI means nothing. And if we look at data today, there is one thing which is very clear. Every single piece of data that comes out of the production system in the world. The data that was used to make this phone, the data that is used to make electric cars. The data that is used to make basically all electronics on the planet today is owned by China. And the consumption data is owned by the United States... China has at least a 10 year lead on India in this space. China has been working on batteries, robots, motors, optics for the last 10 years and we are behind...We would ensure that our banking system is not captured by 2-3 companies that basically do not allow you to build a production system. But our banking system is open, dynamic and accessible to small and medium businesses and to millions and millions of entrepreneurs who want to take part in this revolution. Our foreign policy would take into account this revolution. When we talk to the United States, we would not send our foreign minister to invite our Prime Minister to his coronation...Because if we had a production system and if we were working on these technologies, the American President would come here and invite the Prime Minister."
3:20

#BudgetSession2025 | Lok Sabha LoP and Congress MP Rahul Gandhi says "People talk about AI, but it's important to understand that AI on its own is absolutely meaningless because AI operates on top of data. Without data, AI means nothing. And if we look at data today, there is one thing which is very clear. Every single piece of data that comes out of the production system in the world. The data that was used to make this phone, the data that is used to make electric cars. The data that is used to make basically all electronics on the planet today is owned by China. And the consumption data is owned by the United States... China has at least a 10 year lead on India in this space. China has been working on batteries, robots, motors, optics for the last 10 years and we are behind...We would ensure that our banking system is not captured by 2-3 companies that basically do not allow you to build a production system. But our banking system is open, dynamic and accessible to small and medium businesses and to millions and millions of entrepreneurs who want to take part in this revolution. Our foreign policy would take into account this revolution. When we talk to the United States, we would not send our foreign minister to invite our Prime Minister to his coronation...Because if we had a production system and if we were working on these technologies, the American President would come here and invite the Prime Minister."

ANI

163,581 次观看 • 1 年前

WOW 🚨 Illinois resident says everyone showed up to their City Council meeting to oppose 14 data centers being built She said so many people showed up to oppose it, it took 6 hours with public comments At the end Democrats didn’t care and UNANIMOUSLY VOTED TO APPROVE THEM “I just spent six hours at a city council meeting — where they were discussing the 14 data centers they're gonna build in my city. And public comment went on for hours and hours and hours of people telling them how badly our community does not need this, and how detrimental it will be to the people, the wildlife, the health, the everything, all of the negative impacts of the data centers. We told them, and we told them over and over and over again, and then at the end, after sitting here for six hours, telling them this is not what we want — they unanimously voted yes for the data centers in our face”
2:04

WOW 🚨 Illinois resident says everyone showed up to their City Council meeting to oppose 14 data centers being built She said so many people showed up to oppose it, it took 6 hours with public comments At the end Democrats didn’t care and UNANIMOUSLY VOTED TO APPROVE THEM “I just spent six hours at a city council meeting — where they were discussing the 14 data centers they're gonna build in my city. And public comment went on for hours and hours and hours of people telling them how badly our community does not need this, and how detrimental it will be to the people, the wildlife, the health, the everything, all of the negative impacts of the data centers. We told them, and we told them over and over and over again, and then at the end, after sitting here for six hours, telling them this is not what we want — they unanimously voted yes for the data centers in our face”

Wall Street Apes

1,468,253 次观看 • 6 个月前

🖨️ It took over a year and lots of help from Claude Code but I have now been able to create a real dot matrix printer on the web that prints from Windows 3.11 in your browser: Here's the test page: I couldn't get the LPT1 port to work but COM2 did work (the dial up model already uses COM1), the JS receives the COM2 serial data (just raw print data) and then prints it Next is to make the dot matrix printer better looking (kinda 3d like on pieter dot com) and have the paper also flowing in 3d out of the printer and integrate it on the page My idea was that it might work because we also capture the COM1 data (from the dial up modem) and send it to a Websocket as a dial up connection, so we should be able to capture the COM2 traffic too but I didn't know how, but now I do IT WORKS!! 😊
0:51

🖨️ It took over a year and lots of help from Claude Code but I have now been able to create a real dot matrix printer on the web that prints from Windows 3.11 in your browser: Here's the test page: I couldn't get the LPT1 port to work but COM2 did work (the dial up model already uses COM1), the JS receives the COM2 serial data (just raw print data) and then prints it Next is to make the dot matrix printer better looking (kinda 3d like on pieter dot com) and have the paper also flowing in 3d out of the printer and integrate it on the page My idea was that it might work because we also capture the COM1 data (from the dial up modem) and send it to a Websocket as a dial up connection, so we should be able to capture the COM2 traffic too but I didn't know how, but now I do IT WORKS!! 😊

@levelsio

85,969 次观看 • 3 个月前

For the past 3 months, I've been heads down building and extending our multimodal data capture rig for Physical AI training. What you're seeing here is a single synchronized capture from our system — RGB, depth (stereo IR camera) feed; real-time 3D hand reconstruction, and a 23×20 tactile pressure grid from sensor gloves, all aligned and timestamped together. Getting here took a lot of unglamorous iteration. We reworked the pipeline to deliver clean, gap-free depth at scale after identifying early depth frame drops caused by data throughput bottlenecks in our capture box. We built per-channel calibration models that convert raw resistance signals from our flex sensor gloves into accurate, drift-corrected force values — solving the creep behavior that made a steady grip falsely appear to fade over time. Even though we’ve stayed quiet on socials, word of mouth spread — our rigs have already been deployed in data factories, manufacturing lines, and homes across the US, Europe, China, and Southeast Asia by our customers and ourselves. In the next few days, I'll be sharing more about how we're tackling rights-cleared data capture from the hardware level, and our closed-loop data-model flywheel — using model evaluation results to identify gaps and guide what data our customers need. More soon, stay tuned…
0:26

For the past 3 months, I've been heads down building and extending our multimodal data capture rig for Physical AI training. What you're seeing here is a single synchronized capture from our system — RGB, depth (stereo IR camera) feed; real-time 3D hand reconstruction, and a 23×20 tactile pressure grid from sensor gloves, all aligned and timestamped together. Getting here took a lot of unglamorous iteration. We reworked the pipeline to deliver clean, gap-free depth at scale after identifying early depth frame drops caused by data throughput bottlenecks in our capture box. We built per-channel calibration models that convert raw resistance signals from our flex sensor gloves into accurate, drift-corrected force values — solving the creep behavior that made a steady grip falsely appear to fade over time. Even though we’ve stayed quiet on socials, word of mouth spread — our rigs have already been deployed in data factories, manufacturing lines, and homes across the US, Europe, China, and Southeast Asia by our customers and ourselves. In the next few days, I'll be sharing more about how we're tackling rights-cleared data capture from the hardware level, and our closed-loop data-model flywheel — using model evaluation results to identify gaps and guide what data our customers need. More soon, stay tuned…

Grace Zhang

54,991 次观看 • 3 个月前

We just made the Warp API and MCP completely free. Most HR platforms make you sit through a sales call and a questionnaire just to reach your own data. We think that's backwards. You can pull your entire company and employee data, time off, compensation, payroll, IT, basically anything in Warp. Build your own dashboards. Connect Warp to your internal systems. Run your own automations on top of your data. We built this because our customers are technical and AI-native. They don't want to stare at a dashboard. They want to build on top of their data. Legacy platforms silo it on purpose, and the API, when it exists at all, is a pricey add-on buried behind a sales process. We shipped the API eight weeks ago, and some of our top customers have already built real tools on top of Warp. Today we're removing the last barrier to doing that. If you're an AI-native founder, exec, CIO, or Chief People Officer who's been duct taping this together, the API and MCP are now included free on most plans. On Enterprise, a small fee unlocks the more powerful version with uptime SLAs, webhooks, and audit logs. Your data should be yours. Excited to see what you build with it.
1:23

We just made the Warp API and MCP completely free. Most HR platforms make you sit through a sales call and a questionnaire just to reach your own data. We think that's backwards. You can pull your entire company and employee data, time off, compensation, payroll, IT, basically anything in Warp. Build your own dashboards. Connect Warp to your internal systems. Run your own automations on top of your data. We built this because our customers are technical and AI-native. They don't want to stare at a dashboard. They want to build on top of their data. Legacy platforms silo it on purpose, and the API, when it exists at all, is a pricey add-on buried behind a sales process. We shipped the API eight weeks ago, and some of our top customers have already built real tools on top of Warp. Today we're removing the last barrier to doing that. If you're an AI-native founder, exec, CIO, or Chief People Officer who's been duct taping this together, the API and MCP are now included free on most plans. On Enterprise, a small fee unlocks the more powerful version with uptime SLAs, webhooks, and audit logs. Your data should be yours. Excited to see what you build with it.

Ayush S

730,614 次观看 • 4 个月前

AMAZING NEWS for Geophysics reseachers!!!!! I got a wild idea, and decided to give it a try! Grok build was able to reverse engineer and build a NEW version of the old school Japanese NICT Magnetosphere / Magnetopause model which shows the Earths current magnetic environment in space! The old NICT site was shut down due to "lack of funding" after the great Japan megaquake in 2011, the service ran for a little while after the quake and then went away for good. I was able to get the old site in the wayback machine internet archive, give a link to Grok, with a little direction on where to look for the data they were using, and next then you know Grok has made a full 3D version that pulls data real time ! Now we have a basic working 3D version of the magnetopause , and soon will be coming the animations showing the data over time (we will be able to hit play and watch it bounce around !) This is next level. For real, I'm geeking out!!!!!!
6:03

AMAZING NEWS for Geophysics reseachers!!!!! I got a wild idea, and decided to give it a try! Grok build was able to reverse engineer and build a NEW version of the old school Japanese NICT Magnetosphere / Magnetopause model which shows the Earths current magnetic environment in space! The old NICT site was shut down due to "lack of funding" after the great Japan megaquake in 2011, the service ran for a little while after the quake and then went away for good. I was able to get the old site in the wayback machine internet archive, give a link to Grok, with a little direction on where to look for the data they were using, and next then you know Grok has made a full 3D version that pulls data real time ! Now we have a basic working 3D version of the magnetopause , and soon will be coming the animations showing the data over time (we will be able to hit play and watch it bounce around !) This is next level. For real, I'm geeking out!!!!!!

Dutchsinse

43,914 次观看 • 1 个月前

There are over 90 robot gyms launching in China right now. 100+ robots per center. We are in the robot data gold rush. China's humanoid makers now ship 97% of all humanoids globally. The central government just put $138 billion into a robotics venture fund. And this is only the tip of the iceberg. Robots need an ungodly amount of data to work in the real world. The biggest bottleneck isn't hardware, it's data. So a whole industry appeared overnight to feed them. Welcome to our generation's gold rush: the robot data gold rush. We made a full video on it. We cover why it's such an insane business, how it actually works, the shady stuff (yes, sweatshops for robot data are a thing), and why this might be the completely wrong approach.
17:42

There are over 90 robot gyms launching in China right now. 100+ robots per center. We are in the robot data gold rush. China's humanoid makers now ship 97% of all humanoids globally. The central government just put $138 billion into a robotics venture fund. And this is only the tip of the iceberg. Robots need an ungodly amount of data to work in the real world. The biggest bottleneck isn't hardware, it's data. So a whole industry appeared overnight to feed them. Welcome to our generation's gold rush: the robot data gold rush. We made a full video on it. We cover why it's such an insane business, how it actually works, the shady stuff (yes, sweatshops for robot data are a thing), and why this might be the completely wrong approach.

Andreas Klinger 🦾

13,810 次观看 • 29 天前

New PNAS paper. Historical GDP per capita data is scarce, but data on the places of birth, death, and occupations of famous individuals is abundant. In this paper we estimate the historical GDP per capita of hundreds of regions in Europe and North America using a machine learning model that leveraged data on about 500k famous biographies. Our estimates more-or-less quadruple the availability of historical GDP per capita estimates for the last 700 years. So why use biographies to augment historical GDP per capita data? Biographical data contains information about people who might have contributed directly to economic growth, like James Watt, or that were attracted to wealthy places looking for patrons, like Michelangelo. So we--mainly Philipp (Philipp Koch)--used this data to construct hundreds of features describing each European region. Then, we trained a machine learning model to find the features that explained most of the variance in a cross-validation test, where we split regions multiple times into a training set and a test set. On average, the model explained about 90% of the variance in GDP per capita of the regions it had not seen during training. But we wanted to go further, and Philipp really went to town by looking at different ways to validate our estimates. We found our estimates correlate positively with historical measures of wellbeing, church building activity, urbanization, and body height. We also used these measures to reproduce the basic Atlantic trade result of Acemoglu, Johnson, and Robison and to explore the economic consequences of the famous Lisbon earthquake of 1755. But what I personally loved most about this project, other than working with Philipp Koch and V, is that it shows that we can use machine learning methods not only to explore the future, but the past. There is a bright and growing future in the use of machine learning for economic history. Hope you enjoy the paper and the data. You can find links to the paper and a data exploration tool in the first comment.
3:25

New PNAS paper. Historical GDP per capita data is scarce, but data on the places of birth, death, and occupations of famous individuals is abundant. In this paper we estimate the historical GDP per capita of hundreds of regions in Europe and North America using a machine learning model that leveraged data on about 500k famous biographies. Our estimates more-or-less quadruple the availability of historical GDP per capita estimates for the last 700 years. So why use biographies to augment historical GDP per capita data? Biographical data contains information about people who might have contributed directly to economic growth, like James Watt, or that were attracted to wealthy places looking for patrons, like Michelangelo. So we--mainly Philipp (Philipp Koch)--used this data to construct hundreds of features describing each European region. Then, we trained a machine learning model to find the features that explained most of the variance in a cross-validation test, where we split regions multiple times into a training set and a test set. On average, the model explained about 90% of the variance in GDP per capita of the regions it had not seen during training. But we wanted to go further, and Philipp really went to town by looking at different ways to validate our estimates. We found our estimates correlate positively with historical measures of wellbeing, church building activity, urbanization, and body height. We also used these measures to reproduce the basic Atlantic trade result of Acemoglu, Johnson, and Robison and to explore the economic consequences of the famous Lisbon earthquake of 1755. But what I personally loved most about this project, other than working with Philipp Koch and V, is that it shows that we can use machine learning methods not only to explore the future, but the past. There is a bright and growing future in the use of machine learning for economic history. Hope you enjoy the paper and the data. You can find links to the paper and a data exploration tool in the first comment.

César A. Hidalgo

54,358 次观看 • 2 年前

Georgia Power is trying to push a family off their farm via eminent domain for a data center: “I'm fighting for the survival of my cattle farm. I'm here because a massive data center was approved just a couple of miles from my land — and I'm being hounded by Georgia Power for an easement to build transmission lines through my property for the data center.” “I'm a local farmer, not an industrial developer. These 500-kV lines aren't for me. They are for the data centers that the boards and surrounding counties continue to approve. I have mail from lawyers stacking up on my kitchen table, wanting to take my case because they know my land is being targeted for eminent domain — These easements are permanent. They affect my ability to graze my cattle, they lower my property value, and they destroy the rural character of this county forever. This board makes decisions to approve these massive, massive projects, but it's residents like me, young people trying to build a life here, who pay the price. You're voting to turn our farms into a network of high-voltage wires and noisy industrial buildings. I'm asking you to realize the real-world impacts of your votes. Every time you say yes to a data center, you're saying no to a local farmer. We aren't just numbers on a map. We are the future of the county, and right now you're making that future impossible.“ This project impacts more than 330 private properties. Georgia Power says it will negotiate purchases and easements and, if needed, use eminent domain. They claim it's to strengthen the grid for Georgia's rising energy demand from new data centers. The lines are widely linked to Project Sail — a $17 billion hyperscale campus by Prologis, Atlas with 9 massive buildings totaling up to 4.34 million sq ft on 829 acres, demanding hundreds of megawatts, roughly a small city's power. We cannot allow data centers to be prioritized over farmers.
2:00

Georgia Power is trying to push a family off their farm via eminent domain for a data center: “I'm fighting for the survival of my cattle farm. I'm here because a massive data center was approved just a couple of miles from my land — and I'm being hounded by Georgia Power for an easement to build transmission lines through my property for the data center.” “I'm a local farmer, not an industrial developer. These 500-kV lines aren't for me. They are for the data centers that the boards and surrounding counties continue to approve. I have mail from lawyers stacking up on my kitchen table, wanting to take my case because they know my land is being targeted for eminent domain — These easements are permanent. They affect my ability to graze my cattle, they lower my property value, and they destroy the rural character of this county forever. This board makes decisions to approve these massive, massive projects, but it's residents like me, young people trying to build a life here, who pay the price. You're voting to turn our farms into a network of high-voltage wires and noisy industrial buildings. I'm asking you to realize the real-world impacts of your votes. Every time you say yes to a data center, you're saying no to a local farmer. We aren't just numbers on a map. We are the future of the county, and right now you're making that future impossible.“ This project impacts more than 330 private properties. Georgia Power says it will negotiate purchases and easements and, if needed, use eminent domain. They claim it's to strengthen the grid for Georgia's rising energy demand from new data centers. The lines are widely linked to Project Sail — a $17 billion hyperscale campus by Prologis, Atlas with 9 massive buildings totaling up to 4.34 million sq ft on 829 acres, demanding hundreds of megawatts, roughly a small city's power. We cannot allow data centers to be prioritized over farmers.

redpillbot

60,418 次观看 • 4 个月前

Today we're launching Hebbian Robotics (YC S26) Hebbian Robotics . We build APIs for evaluating data quality, without training a robotics model. Data vendors use our API to search and analyze Physical AI data at scale, getting on-demand quality signals and metrics without managing infrastructure. High-quality data is the bottleneck to the next frontier of intelligence in robotics. Buyers want data they can trust. Data providers are focused on operational excellence. They don't have the bandwidth to understand the quality of their data at scale. We're fixing this. Vendors use our APIs to understand the impact of their decisions on their data quality: - Buyers demanding per-collector hour limits? Prove compliance while we stay ahead of the workarounds. - Changed your collection SOPs? Monitor for quality drift and regression. - Suspect duplicate samples? Verify with a single API call. Before Hebbian, I trained AI models for industrial cleaning robots, and Kingston Kuan spent years building large-scale infrastructure at Jane Street.
2:17

Today we're launching Hebbian Robotics (YC S26) Hebbian Robotics . We build APIs for evaluating data quality, without training a robotics model. Data vendors use our API to search and analyze Physical AI data at scale, getting on-demand quality signals and metrics without managing infrastructure. High-quality data is the bottleneck to the next frontier of intelligence in robotics. Buyers want data they can trust. Data providers are focused on operational excellence. They don't have the bandwidth to understand the quality of their data at scale. We're fixing this. Vendors use our APIs to understand the impact of their decisions on their data quality: - Buyers demanding per-collector hour limits? Prove compliance while we stay ahead of the workarounds. - Changed your collection SOPs? Monitor for quality drift and regression. - Suspect duplicate samples? Verify with a single API call. Before Hebbian, I trained AI models for industrial cleaning robots, and Kingston Kuan spent years building large-scale infrastructure at Jane Street.

Brandon Ong

166,693 次观看 • 1 个月前