Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

Sharing some exciting DYNA-1 result: zero-shot environment generalization We put DYNA-1 under test in a completely different environment from our training distribution – with an entirely different background (Dyna Robotics banner) and metal table. The table has a reflective and smooth surface, creating a wildly different visual appearance as... show more

Jason Ma

20,654 subscribers

49,008 Aufrufe • vor 1 Jahr •via X (Twitter)

Wissenschaft & Technologie Bildung

Anya Rossi• Live Now

Private livecam show

10 Kommentare

Profilbild von Jason Ma

Jason Mavor 1 Jahr

Original thread:

Profilbild von MrRobotics

MrRoboticsvor 1 Jahr

@DynaRobotics why is it 10X and doesn't look impressive as the first demo?

Profilbild von Hiro Protagonist

Hiro Protagonistvor 1 Jahr

@DynaRobotics 10x speed so it's still basically useless for real life applications. Robotics amateur hour.

Profilbild von pfung

pfungvor 1 Jahr

@DynaRobotics nice robustness!

Profilbild von Ming Qin

Ming Qinvor 1 Jahr

@DynaRobotics The table can even reflect the robot itself, and it doesn’t seem surprised at all

Profilbild von atharva

atharvavor 1 Jahr

@DynaRobotics insane

Profilbild von Max von Wolff

Max von Wolffvor 1 Jahr

@DynaRobotics Impressive!

Profilbild von Humanoid Pulse

Humanoid Pulsevor 1 Jahr

@DynaRobotics wow- there is something sacinating in that - can watch for hours🤓

Profilbild von Peter Christie

Peter Christievor 1 Jahr

@DynaRobotics Nope….

Profilbild von Michael Cho - Rbt/Acc

Michael Cho - Rbt/Accvor 1 Jahr

@DynaRobotics Impressive stuff!

Ähnliche Videos

Introducing Dynamism v1 (DYNA-1) by Dyna Robotics – the first robot foundation model built for round-the-clock, high-throughput dexterous autonomy. Here is a time-lapse video of our model autonomously folding 850+ napkins in a span of 24 hours with • 99.4% success rate — zero human intervention • 60% human throughput speed • 4.3/5 quality ratings (set by the client) A thread on our motivation, insights and results:

Introducing Dynamism v1 (DYNA-1) by Dyna Robotics – the first robot foundation model built for round-the-clock, high-throughput dexterous autonomy. Here is a time-lapse video of our model autonomously folding 850+ napkins in a span of 24 hours with • 99.4% success rate — zero human intervention • 60% human throughput speed • 4.3/5 quality ratings (set by the client) A thread on our motivation, insights and results:

Jason Ma

523,118 Aufrufe • vor 1 Jahr

How well do egocentric human videos scale? Apparently, really well. Dyna Robotics has introduced Dyna-2, a world-action model (WAM) pre-trained on more than 1 million hours of real egocentric human video. It learns by jointly predicting the next video frames and the actions, and that video prediction (world modeling) is what carries learning from humans to robots. • More human video predictably improves action prediction. • Task accuracy improves across embodiments (bimanual arms and a humanoid) with zero robot data in pre-training. • Post-trained on just a few hours of robot data, mean task performance rose 20% to 53% as human pre-training scaled 1k to 1M hours. WAM vs VLA, matched apple-to-apple: the WAM hit 1.55x the VLA's success rate and won 65% of head-to-heads. Deployed zero-shot at unseen customer sites, it passed 87% vs the VLA's 46%.

How well do egocentric human videos scale? Apparently, really well. Dyna Robotics has introduced Dyna-2, a world-action model (WAM) pre-trained on more than 1 million hours of real egocentric human video. It learns by jointly predicting the next video frames and the actions, and that video prediction (world modeling) is what carries learning from humans to robots. • More human video predictably improves action prediction. • Task accuracy improves across embodiments (bimanual arms and a humanoid) with zero robot data in pre-training. • Post-trained on just a few hours of robot data, mean task performance rose 20% to 53% as human pre-training scaled 1k to 1M hours. WAM vs VLA, matched apple-to-apple: the WAM hit 1.55x the VLA's success rate and won 65% of head-to-heads. Deployed zero-shot at unseen customer sites, it passed 87% vs the VLA's 46%.

The Humanoid Hub

15,837 Aufrufe • vor 1 Monat

JUST IN: Dyna Robotics just published one of the most important research papers in robotics this year. It could fundamentally change how robot foundation models are trained. A scaling law that transfers from human video to robot performance. Dyna-2 is out and it's 🔥 Here's what that means in plain terms. Dyna-2 was pre-trained on ONE MILLION hours of egocentric human video, 170 years of continuous human experience, cooking, folding, assembling, cleaning. And as that human data scaled, robot performance improved. Predictably. Monotonically. Across 39 tasks on two different robot embodiments the model had never seen. → 1,000 hours pre-training → 20% normalised task performance → 10,000 hours → 28% → 100,000 hours → 45% → 1,000,000 hours → 53% Human video exists at effectively unlimited scale. Every cook, every factory worker, every craftsperson wearing a camera is generating training data for future robots. But the finding that stunned even the researchers, world modeling is what makes the transfer work. A model trained to predict future video AND actions massively outperforms one trained on actions alone. Video is the new scaling axis for robotics. One more jaw-dropping data point. 13 minutes of teleoperation data was enough to fine-tune Dyna-2 to open a bottle cap using two five-fingered robot hands. The robots are coming, and they're learning from us directly :D Read more here: Congrats Jason Ma and team! ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

JUST IN: Dyna Robotics just published one of the most important research papers in robotics this year. It could fundamentally change how robot foundation models are trained. A scaling law that transfers from human video to robot performance. Dyna-2 is out and it's 🔥 Here's what that means in plain terms. Dyna-2 was pre-trained on ONE MILLION hours of egocentric human video, 170 years of continuous human experience, cooking, folding, assembling, cleaning. And as that human data scaled, robot performance improved. Predictably. Monotonically. Across 39 tasks on two different robot embodiments the model had never seen. → 1,000 hours pre-training → 20% normalised task performance → 10,000 hours → 28% → 100,000 hours → 45% → 1,000,000 hours → 53% Human video exists at effectively unlimited scale. Every cook, every factory worker, every craftsperson wearing a camera is generating training data for future robots. But the finding that stunned even the researchers, world modeling is what makes the transfer work. A model trained to predict future video AND actions massively outperforms one trained on actions alone. Video is the new scaling axis for robotics. One more jaw-dropping data point. 13 minutes of teleoperation data was enough to fine-tune Dyna-2 to open a bottle cap using two five-fingered robot hands. The robots are coming, and they're learning from us directly :D Read more here: Congrats Jason Ma and team! ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

23,576 Aufrufe • vor 1 Monat

A beautiful day with my daughter Rafif in the neighborhoods of Al-Mawasi, Khan Younis. We try, as much as we can, to hold on to moments of joy and to keep going, waiting for our simple right to life, safety, and stability. A different environment from the harsh reality we live in here. Rafif is a child of this world, and her right—and the right of all children in Gaza—must return one day.

A beautiful day with my daughter Rafif in the neighborhoods of Al-Mawasi, Khan Younis. We try, as much as we can, to hold on to moments of joy and to keep going, waiting for our simple right to life, safety, and stability. A different environment from the harsh reality we live in here. Rafif is a child of this world, and her right—and the right of all children in Gaza—must return one day.

Mohammed jawad 🇵🇸

13,636 Aufrufe • vor 7 Monaten

OpenAI researcher David Holtz reveals workers who don't adapt their prompting to newer models leave 50% of the alpha on the table because half of AI's value comes from how you interact with it: "We ran a lab experiment. We give people the same task to do, but we give them models of different capabilities. How much of the improvement when we give people more capable models comes from the fact that we gave them a better model vs the fact that they actually update and adapt the way they prompt and interact with these models?" "For tasks where precision matters, where you have some output you're trying to get and you're trying to steer the model, which is different from a creative task like brainstorming." "For these steering-based tasks, when you give people a better model, about 50% of the improvement we see in how well they're able to complete the task comes from changes in how they interact with the model." "If you just kept the way you interact with the model static and just made the capabilities better and better, you'd leave 50% of the alpha basically on the table." OpenAI

OpenAI researcher David Holtz reveals workers who don't adapt their prompting to newer models leave 50% of the alpha on the table because half of AI's value comes from how you interact with it: "We ran a lab experiment. We give people the same task to do, but we give them models of different capabilities. How much of the improvement when we give people more capable models comes from the fact that we gave them a better model vs the fact that they actually update and adapt the way they prompt and interact with these models?" "For tasks where precision matters, where you have some output you're trying to get and you're trying to steer the model, which is different from a creative task like brainstorming." "For these steering-based tasks, when you give people a better model, about 50% of the improvement we see in how well they're able to complete the task comes from changes in how they interact with the model." "If you just kept the way you interact with the model static and just made the capabilities better and better, you'd leave 50% of the alpha basically on the table." OpenAI

MTS

15,590 Aufrufe • vor 1 Monat

Jason Ma reveals the first true scaling law in robotics: robots trained on 1 million hours of first-person human video predictably get better without ever seeing robot data "Yesterday we announced our new flagship robot foundation model called Dyna-2. It's very significant for the entire field for many different reasons. First of all, it's the first robot foundation model trained on at least one million hours of data. That in itself is a very challenging infrastructure challenge." "What we demonstrated is that even if the one million hours of data is entirely first-person video of humans doing manipulation tasks, we actually see scaling law for transferring to robot embodiments that the model has never seen before." "By training on more human data, we actually see predictable performance improvement on robots. That's a big deal because the biggest challenge facing robot foundation models is that we don't have enough data." "Unlike language models, there are just not enough robotics data out there, and it's very difficult to collect these robot data. Only by passively observing humans doing things can we scale this kind of data, and this is the first time we showed there is a lot of improvement that can transfer." Jason Ma Dyna Robotics

Jason Ma reveals the first true scaling law in robotics: robots trained on 1 million hours of first-person human video predictably get better without ever seeing robot data "Yesterday we announced our new flagship robot foundation model called Dyna-2. It's very significant for the entire field for many different reasons. First of all, it's the first robot foundation model trained on at least one million hours of data. That in itself is a very challenging infrastructure challenge." "What we demonstrated is that even if the one million hours of data is entirely first-person video of humans doing manipulation tasks, we actually see scaling law for transferring to robot embodiments that the model has never seen before." "By training on more human data, we actually see predictable performance improvement on robots. That's a big deal because the biggest challenge facing robot foundation models is that we don't have enough data." "Unlike language models, there are just not enough robotics data out there, and it's very difficult to collect these robot data. Only by passively observing humans doing things can we scale this kind of data, and this is the first time we showed there is a lot of improvement that can transfer." Jason Ma Dyna Robotics

MTS

17,637 Aufrufe • vor 1 Monat

Here’s How 1 McDonald’s Hamburger Contains 1300 Strands Of Different DNA. This Is The Result Of Going From Small Farms To Mega Corporations “We're pooling the bacteria from a thousand different animals” “Thousands of animals” contribute “to a single hamburger.” “This is a combo bin of meat. The meat in these boxes comes from many different animals and many different places. Some is even from overseas. All of it is ground together to make the juicy hamburger Americans love. Hamburger used to be the scraps left over from butchering just one cow.” But a significant change has occurred. Now parts of many cows are blended together by the ton. ‌ If we take the meat from one animal, grind it up and make ground beef from just one animal, we're including only the bacteria from one animal. But if we take the meat from a thousand different animals and grind that together, we're pooling the bacteria from a thousand different animals as well. So explain this to me, how many different cows would be in one burger? Well, I suspect that there are hundreds or even thousands of animals that have contributed to a single hamburger.”

Here’s How 1 McDonald’s Hamburger Contains 1300 Strands Of Different DNA. This Is The Result Of Going From Small Farms To Mega Corporations “We're pooling the bacteria from a thousand different animals” “Thousands of animals” contribute “to a single hamburger.” “This is a combo bin of meat. The meat in these boxes comes from many different animals and many different places. Some is even from overseas. All of it is ground together to make the juicy hamburger Americans love. Hamburger used to be the scraps left over from butchering just one cow.” But a significant change has occurred. Now parts of many cows are blended together by the ton. ‌ If we take the meat from one animal, grind it up and make ground beef from just one animal, we're including only the bacteria from one animal. But if we take the meat from a thousand different animals and grind that together, we're pooling the bacteria from a thousand different animals as well. So explain this to me, how many different cows would be in one burger? Well, I suspect that there are hundreds or even thousands of animals that have contributed to a single hamburger.”

Wall Street Apes

3,114,375 Aufrufe • vor 2 Jahren

A question I kept returning to was how soon robots will be ubiquitous in our daily lives. Sergey uses self-driving as an analogy, but explains why robotics may not follow the same path because the cost of mistakes is very different. Self-driving took over a decade and is still in its nascency, in part because even rare mistakes are unacceptable. Robots may deploy earlier, and improve faster, because some mistakes are tolerable. Breaking a dish is different from causing an accident.

A question I kept returning to was how soon robots will be ubiquitous in our daily lives. Sergey uses self-driving as an analogy, but explains why robotics may not follow the same path because the cost of mistakes is very different. Self-driving took over a decade and is still in its nascency, in part because even rare mistakes are unacceptable. Robots may deploy earlier, and improve faster, because some mistakes are tolerable. Breaking a dish is different from causing an accident.

Patrick OShaughnessy

57,780 Aufrufe • vor 5 Monaten

Curious whether video generation models (like #SORA) qualify as world models? We conduct a systematic study to answer this question by investigating whether a video gen model is able to learn physical laws. Three are three key messages to take home: 1⃣The model generalises perfectly for in-distribution data, but fails to do out-of-distribution generalization. For combinatorial scenarios, scaling law is observed. 2⃣The models fail to abstract general rules and instead tries to mimic the closest training example. 3⃣The model prioritizes different attributes when referencing training data: color > size > velocity > shape. This work is a joint effort with our outstanding intern Yang Yue. Paper: Webpage:

Curious whether video generation models (like #SORA) qualify as world models? We conduct a systematic study to answer this question by investigating whether a video gen model is able to learn physical laws. Three are three key messages to take home: 1⃣The model generalises perfectly for in-distribution data, but fails to do out-of-distribution generalization. For combinatorial scenarios, scaling law is observed. 2⃣The models fail to abstract general rules and instead tries to mimic the closest training example. 3⃣The model prioritizes different attributes when referencing training data: color > size > velocity > shape. This work is a joint effort with our outstanding intern Yang Yue. Paper: Webpage:

Bingyi Kang

606,773 Aufrufe • vor 1 Jahr

Physics feels stable, predictable, and well-behaved largely because we learned it in 3D. That comfort hides a trap. In 1907, Paul Ehrenfest pointed out something unsettling. If you take the laws we treat as fundamental and transplant them into a different number of spatial dimensions, they often stop working the way we expect. Not just numerically different but qualitatively different. The issue isn’t the force law by itself but it's geometry. Gauss’s law ties inverse-square forces to the surface area of spheres, and sphere geometry depends on dimension. Change the dimension, and the same-looking force produces a different potential, a different balance of attraction and inertia, and a different fate for motion.

Physics feels stable, predictable, and well-behaved largely because we learned it in 3D. That comfort hides a trap. In 1907, Paul Ehrenfest pointed out something unsettling. If you take the laws we treat as fundamental and transplant them into a different number of spatial dimensions, they often stop working the way we expect. Not just numerically different but qualitatively different. The issue isn’t the force law by itself but it's geometry. Gauss’s law ties inverse-square forces to the surface area of spheres, and sphere geometry depends on dimension. Change the dimension, and the same-looking force produces a different potential, a different balance of attraction and inertia, and a different fate for motion.

Mathelirium

30,521 Aufrufe • vor 5 Monaten

Robots don’t learn the physical world from clean datasets. They learn from messy decisions, different movements, and thousands of possible ways a task can go wrong. That’s why I think the data problem in Physical AI is much bigger than simply collecting more examples. A robot can look impressive in a controlled demonstration and still struggle when the environment changes. A different angle. A different object. A slightly different position. A movement that doesn’t go exactly as expected. Humans handle these variations naturally because we have years of physical experience to draw from. Robots don’t. They need enormous amounts of training experience to learn how to perceive situations, choose actions, recover from mistakes, and repeat successful behavior. This is where Axis Robotics is taking an interesting approach. Instead of treating simulation as just a place to test robots, Axis is turning it into a data-generation environment for physical intelligence. More than 200K people are now contributing demonstrations in simulation, creating examples of how tasks can be performed. But the important part isn’t simply the number of contributors. It’s what happens to those demonstrations afterward. The interactions are recorded, validated, and written to Base, creating a verifiable trail around the data being generated. That gives the whole process a different structure: Humans demonstrate → simulations capture → data gets validated → training signals accumulate → robots get more experience. And that feedback loop is what interests me. Because scaling Physical AI may not come from finding one perfect algorithm. It may come from building a system capable of producing millions of useful experiences that algorithms can actually learn from. The robot is only as good as the experience available to train it. And if Axis can keep scaling the quality and volume of that experience, it could become one of the important infrastructure layers behind the next generation of capable robots. The race in Physical AI isn’t only about building smarter models. It’s about giving those models enough real-world experience to become smart in the first place. Axis Robotics.

Robots don’t learn the physical world from clean datasets. They learn from messy decisions, different movements, and thousands of possible ways a task can go wrong. That’s why I think the data problem in Physical AI is much bigger than simply collecting more examples. A robot can look impressive in a controlled demonstration and still struggle when the environment changes. A different angle. A different object. A slightly different position. A movement that doesn’t go exactly as expected. Humans handle these variations naturally because we have years of physical experience to draw from. Robots don’t. They need enormous amounts of training experience to learn how to perceive situations, choose actions, recover from mistakes, and repeat successful behavior. This is where Axis Robotics is taking an interesting approach. Instead of treating simulation as just a place to test robots, Axis is turning it into a data-generation environment for physical intelligence. More than 200K people are now contributing demonstrations in simulation, creating examples of how tasks can be performed. But the important part isn’t simply the number of contributors. It’s what happens to those demonstrations afterward. The interactions are recorded, validated, and written to Base, creating a verifiable trail around the data being generated. That gives the whole process a different structure: Humans demonstrate → simulations capture → data gets validated → training signals accumulate → robots get more experience. And that feedback loop is what interests me. Because scaling Physical AI may not come from finding one perfect algorithm. It may come from building a system capable of producing millions of useful experiences that algorithms can actually learn from. The robot is only as good as the experience available to train it. And if Axis can keep scaling the quality and volume of that experience, it could become one of the important infrastructure layers behind the next generation of capable robots. The race in Physical AI isn’t only about building smarter models. It’s about giving those models enough real-world experience to become smart in the first place. Axis Robotics.

ABBA

90,854 Aufrufe • vor 2 Tagen

Sincitium is finally here. We are pleased to present our latest piece: a concept trailer created specifically for the Runway Big Pitch Contest. For this project, we wanted to explore a completely different aesthetic from our usual studio style, and this film is the result of that experimentation. We hope you enjoy it as much as we enjoyed the creative process. Produced by: Contanimation Directed by: Javier De La Chica and Guillermo Miranda Art Direction: Javier De La Chica Editing: Guillermo Miranda Voices: Juan Rabadán #runwaybigpitchcontest

Sincitium is finally here. We are pleased to present our latest piece: a concept trailer created specifically for the Runway Big Pitch Contest. For this project, we wanted to explore a completely different aesthetic from our usual studio style, and this film is the result of that experimentation. We hope you enjoy it as much as we enjoyed the creative process. Produced by: Contanimation Directed by: Javier De La Chica and Guillermo Miranda Art Direction: Javier De La Chica Editing: Guillermo Miranda Voices: Juan Rabadán #runwaybigpitchcontest

Contanimation

3,077,040 Aufrufe • vor 4 Monaten

The holy grail of robotics is to be able to perform previously-unseen, out-of-distribution manipulation tasks “zero shot” in a new environment. NovaFlow proposes an approach which (1) generates a video, (2) computes predicted flow — how points move through the scene — and (3) uses this flow as an objective to generate a motion. Using this procedure, NovaFlow generates motions in unseen scenes, for unseen tasks, and can transfer across embodiments. To learn more, we are joined by Hongyu Li and Jiahui Fu from RAI. Watch Episode #63 of RoboPapers with Chris Paxton and Michael Cho - Rbt/Acc now to learn more!

The holy grail of robotics is to be able to perform previously-unseen, out-of-distribution manipulation tasks “zero shot” in a new environment. NovaFlow proposes an approach which (1) generates a video, (2) computes predicted flow — how points move through the scene — and (3) uses this flow as an objective to generate a motion. Using this procedure, NovaFlow generates motions in unseen scenes, for unseen tasks, and can transfer across embodiments. To learn more, we are joined by Hongyu Li and Jiahui Fu from RAI. Watch Episode #63 of RoboPapers with Chris Paxton and Michael Cho - Rbt/Acc now to learn more!

RoboPapers

21,434 Aufrufe • vor 6 Monaten

Shout-out to the artist community We love art, images, and ideas that make the world feel different, even strange at times. As a team, we are always curious about new art and enjoy sharing the gems we stumble upon with our community. From photography to AI and code-based art, here are some of the works we have loved over the past few weeks. We hope you enjoy them as much as we do! ↓

Shout-out to the artist community We love art, images, and ideas that make the world feel different, even strange at times. As a team, we are always curious about new art and enjoy sharing the gems we stumble upon with our community. From photography to AI and code-based art, here are some of the works we have loved over the past few weeks. We hope you enjoy them as much as we do! ↓

Fellowship

28,817 Aufrufe • vor 4 Monaten

TEDRA Text-based Editing of Dynamic and Photoreal Actors discuss: Over the past years, significant progress has been made in creating photorealistic and drivable 3D avatars solely from videos of real humans. However, a core remaining challenge is the fine-grained and user-friendly editing of clothing styles by means of textual descriptions. To this end, we present TEDRA, the first method allowing text-based edits of an avatar, which maintains the avatar's high fidelity, space-time coherency, as well as dynamics, and enables skeletal pose and view control. We begin by training a model to create a controllable and high-fidelity digital replica of the real actor. Next, we personalize a pretrained generative diffusion model by fine-tuning it on various frames of the real character captured from different camera angles, ensuring the digital representation faithfully captures the dynamics and movements of the real person. This two-stage process lays the foundation for our approach to dynamic human avatar editing. Utilizing this personalized diffusion model, we modify the dynamic avatar based on a provided text prompt using our Personalized Normal Aligned Score Distillation Sampling (PNA-SDS) within a model-based guidance framework. Additionally, we propose a time step annealing strategy to ensure high-quality edits. Our results demonstrate a clear improvement over prior work in functionality and visual quality.

AK

17,302 Aufrufe • vor 2 Jahren

A HISTORIC MOMENT FOR BUZSTOPBOYS! 🚀♻️ FROM PLASTIC WASTE TO PURPOSE. Today, we proudly celebrate a milestone that once seemed like a dream. Buzstopboys has successfully produced its very first recycled product from plastic waste collected from our environment. This is more than a product. It is proof that with vision, commitment, and collective action, waste can become opportunity and pollution can become possibility. POWERED ENTIRELY BY VOLUNTEERS. Every cleanup, every bag of plastic collected, every hour invested, and every step that led to this achievement has been driven by people who simply chose to serve. No one was paid to make this happen. It is the result of selfless volunteers who believe that protecting our environment is everyone's responsibility. THIS IS ONLY THE BEGINNING. Today, we celebrate a recycled product. Tomorrow, we build a movement that transforms communities, creates opportunities, and inspires a new generation to see waste not as an end, but as the beginning of something valuable. To every volunteer, partner, supporter, and believer in our mission-this victory belongs to you. WE DON'T JUST CLEAN OUR ENVIRONMENT. WE CREATE ITS FUTURE. The journey continues.

A HISTORIC MOMENT FOR BUZSTOPBOYS! 🚀♻️ FROM PLASTIC WASTE TO PURPOSE. Today, we proudly celebrate a milestone that once seemed like a dream. Buzstopboys has successfully produced its very first recycled product from plastic waste collected from our environment. This is more than a product. It is proof that with vision, commitment, and collective action, waste can become opportunity and pollution can become possibility. POWERED ENTIRELY BY VOLUNTEERS. Every cleanup, every bag of plastic collected, every hour invested, and every step that led to this achievement has been driven by people who simply chose to serve. No one was paid to make this happen. It is the result of selfless volunteers who believe that protecting our environment is everyone's responsibility. THIS IS ONLY THE BEGINNING. Today, we celebrate a recycled product. Tomorrow, we build a movement that transforms communities, creates opportunities, and inspires a new generation to see waste not as an end, but as the beginning of something valuable. To every volunteer, partner, supporter, and believer in our mission-this victory belongs to you. WE DON'T JUST CLEAN OUR ENVIRONMENT. WE CREATE ITS FUTURE. The journey continues.

Buzstopboys

46,514 Aufrufe • vor 1 Monat

The co-founders of Flapping Airplanes call the current RL paradigm for model training "environment slop." They explain: "The reinforcement paradigms of today are shockingly inefficient. You don't really get much generalization across tasks, you teach a model through one kind of learning and then you teach it the next one. It's kind of like whack-a-mole. We look at this and think it's kind of crazy. The next paradigm of AI will not be environment slop." "Human level intelligence is not the ceiling, it is merely the floor on what is possible. If you can train models with vastly less data and possibly more compute in very different ways, what is going to happen? We actually don't know. But I do think they'll be different and weird and they'll have interesting capabilities that we'll find really valuable ways to use."

The co-founders of Flapping Airplanes call the current RL paradigm for model training "environment slop." They explain: "The reinforcement paradigms of today are shockingly inefficient. You don't really get much generalization across tasks, you teach a model through one kind of learning and then you teach it the next one. It's kind of like whack-a-mole. We look at this and think it's kind of crazy. The next paradigm of AI will not be environment slop." "Human level intelligence is not the ceiling, it is merely the floor on what is possible. If you can train models with vastly less data and possibly more compute in very different ways, what is going to happen? We actually don't know. But I do think they'll be different and weird and they'll have interesting capabilities that we'll find really valuable ways to use."

TBPN

85,358 Aufrufe • vor 7 Monaten