Tag Archives: better
#438998 Foam Sword Fencing With a PR2 Is the ...
Most of what we cover in the Human Robot Interaction (HRI) space involves collaboration, because collaborative interactions tend to be productive, positive, and happy. Yay! But sometimes, collaboration is not what you want. Sometimes, you want competition.
Competition between humans and robots doesn’t have to be a bad thing, in the same way that competition between humans and humans doesn’t have to be a bad thing. There are all kinds of scenarios in which humans respond favorably to competition, and exercise is an obvious example.
Studies have shown that humans can perform significantly better when they’re exercising competitively as opposed to when they’re exercising individually. And while researchers have looked at whether robots can be effective exercise coaches (they can be), there hasn’t been a lot of exploration of physical robots actually competing directly with humans. Roboticists from the University of Washington decided to put adversarial exercise robots to the test, and they did it by giving a PR2 a giant foam sword. Awesome.
This exercise game matches a PR2 with a human in a zero-sum competitive fencing game with foam swords. Expecting the PR2 to actually be a competitive fencer isn’t realistic because, like, it’s a PR2. Instead, the objective of the game is for the human to keep their foam sword within a target area near the PR2 while also avoiding the PR2’s low-key sword-waving. A VR system allows the user to see the target area, while also giving the system a way to track the user’s location and pose.
Looks like fun, right? It’s also exercise, at least in the sense that the user’s heart rate nearly doubled over their resting heart rate during the highest scoring game. This is super preliminary research, though, and there’s still a lot of work to do. It’ll be important to figure out how skilled a competitive robot should be in order to keep providing a reasonable challenge to a human who gradually improves over time, while also being careful to avoid generating any negative reactions. For example, the robot should probably not beat you over the head with its foam sword, even if that’s a highly effective strategy for getting your heart rate up.
Competitive Physical Human-Robot Game Play, by Boling Yang, Xiangyu Xie, Golnaz Habibi, and Joshua R. Smith from the University of Washington and MIT, was presented as a late-breaking report at the ACM/IEEE International Conference on Human-Robot Interaction. Continue reading
Posted in Human Robots
#438982 Quantum Computing and Reinforcement ...
Deep reinforcement learning is having a superstar moment.
Powering smarter robots. Simulating human neural networks. Trouncing physicians at medical diagnoses and crushing humanity’s best gamers at Go and Atari. While far from achieving the flexible, quick thinking that comes naturally to humans, this powerful machine learning idea seems unstoppable as a harbinger of better thinking machines.
Except there’s a massive roadblock: they take forever to run. Because the concept behind these algorithms is based on trial and error, a reinforcement learning AI “agent” only learns after being rewarded for its correct decisions. For complex problems, the time it takes an AI agent to try and fail to learn a solution can quickly become untenable.
But what if you could try multiple solutions at once?
This week, an international collaboration led by Dr. Philip Walther at the University of Vienna took the “classic” concept of reinforcement learning and gave it a quantum spin. They designed a hybrid AI that relies on both quantum and run-of-the-mill classic computing, and showed that—thanks to quantum quirkiness—it could simultaneously screen a handful of different ways to solve a problem.
The result is a reinforcement learning AI that learned over 60 percent faster than its non-quantum-enabled peers. This is one of the first tests that shows adding quantum computing can speed up the actual learning process of an AI agent, the authors explained.
Although only challenged with a “toy problem” in the study, the hybrid AI, once scaled, could impact real-world problems such as building an efficient quantum internet. The setup “could readily be integrated within future large-scale quantum communication networks,” the authors wrote.
The Bottleneck
Learning from trial and error comes intuitively to our brains.
Say you’re trying to navigate a new convoluted campground without a map. The goal is to get from the communal bathroom back to your campsite. Dead ends and confusing loops abound. We tackle the problem by deciding to turn either left or right at every branch in the road. One will get us closer to the goal; the other leads to a half hour of walking in circles. Eventually, our brain chemistry rewards correct decisions, so we gradually learn the correct route. (If you’re wondering…yeah, true story.)
Reinforcement learning AI agents operate in a similar trial-and-error way. As a problem becomes more complex, the number—and time—of each trial also skyrockets.
“Even in a moderately realistic environment, it may simply take too long to rationally respond to a given situation,” explained study author Dr. Hans Briegel at the Universität Innsbruck in Austria, who previously led efforts to speed up AI decision-making using quantum mechanics. If there’s pressure that allows “only a certain time for a response, an agent may then be unable to cope with the situation and to learn at all,” he wrote.
Many attempts have tried speeding up reinforcement learning. Giving the AI agent a short-term “memory.” Tapping into neuromorphic computing, which better resembles the brain. In 2014, Briegel and colleagues showed that a “quantum brain” of sorts can help propel an AI agent’s decision-making process after learning. But speeding up the learning process itself has eluded our best attempts.
The Hybrid AI
The new study went straight for that previously untenable jugular.
The team’s key insight was to tap into the best of both worlds—quantum and classical computing. Rather than building an entire reinforcement learning system using quantum mechanics, they turned to a hybrid approach that could prove to be more practical. Here, the AI agent uses quantum weirdness as it’s trying out new approaches—the “trial” in trial and error. The system then passes the baton to a classical computer to give the AI its reward—or not—based on its performance.
At the heart of the quantum “trial” process is a quirk called superposition. Stay with me. Our computers are powered by electrons, which can represent only two states—0 or 1. Quantum mechanics is far weirder, in that photons (particles of light) can simultaneously be both 0 and 1, with a slightly different probability of “leaning towards” one or the other.
This noncommittal oddity is part of what makes quantum computing so powerful. Take our reinforcement learning example of navigating a new campsite. In our classic world, we—and our AI—need to decide between turning left or right at an intersection. In a quantum setup, however, the AI can (in a sense) turn left and right at the same time. So when searching for the correct path back to home base, the quantum system has a leg up in that it can simultaneously explore multiple routes, making it far faster than conventional, consecutive trail and error.
“As a consequence, an agent that can explore its environment in superposition will learn significantly faster than its classical counterpart,” said Briegel.
It’s not all theory. To test out their idea, the team turned to a programmable chip called a nanophotonic processor. Think of it as a CPU-like computer chip, but it processes particles of light—photons—rather than electricity. These light-powered chips have been a long time in the making. Back in 2017, for example, a team from MIT built a fully optical neural network into an optical chip to bolster deep learning.
The chips aren’t all that exotic. Nanophotonic processors act kind of like our eyeglasses, which can carry out complex calculations that transform light that passes through them. In the glasses case, they let people see better. For a light-based computer chip, it allows computation. Rather than using electrical cables, the chips use “wave guides” to shuttle photons and perform calculations based on their interactions.
The “error” or “reward” part of the new hardware comes from a classical computer. The nanophotonic processor is coupled to a traditional computer, where the latter provides the quantum circuit with feedback—that is, whether to reward a solution or not. This setup, the team explains, allows them to more objectively judge any speed-ups in learning in real time.
In this way, a hybrid reinforcement learning agent alternates between quantum and classical computing, trying out ideas in wibbly-wobbly “multiverse” land while obtaining feedback in grounded, classic physics “normality.”
A Quantum Boost
In simulations using 10,000 AI agents and actual experimental data from 165 trials, the hybrid approach, when challenged with a more complex problem, showed a clear leg up.
The key word is “complex.” The team found that if an AI agent has a high chance of figuring out the solution anyway—as for a simple problem—then classical computing works pretty well. The quantum advantage blossoms when the task becomes more complex or difficult, allowing quantum mechanics to fully flex its superposition muscles. For these problems, the hybrid AI was 63 percent faster at learning a solution compared to traditional reinforcement learning, decreasing its learning effort from 270 guesses to 100.
Now that scientists have shown a quantum boost for reinforcement learning speeds, the race for next-generation computing is even more lit. Photonics hardware required for long-range light-based communications is rapidly shrinking, while improving signal quality. The partial-quantum setup could “aid specifically in problems where frequent search is needed, for example, network routing problems” that’s prevalent for a smooth-running internet, the authors wrote. With a quantum boost, reinforcement learning may be able to tackle far more complex problems—those in the real world—than currently possible.
“We are just at the beginning of understanding the possibilities of quantum artificial intelligence,” said lead author Walther.
Image Credit: Oleg Gamulinskiy from Pixabay Continue reading
Posted in Human Robots
#438807 Visible Touch: How Cameras Can Help ...
The dawn of the robot revolution is already here, and it is not the dystopian nightmare we imagined. Instead, it comes in the form of social robots: Autonomous robots in homes and schools, offices and public spaces, able to interact with humans and other robots in a socially acceptable, human-perceptible way to resolve tasks related to core human needs.
To design social robots that “understand” humans, robotics scientists are delving into the psychology of human communication. Researchers from Cornell University posit that embedding the sense of touch in social robots could teach them to detect physical interactions and gestures. They describe a way of doing so by relying not on touch but on vision.
A USB camera inside the robot captures shadows of hand gestures on the robot’s surface and classifies them with machine-learning software. They call this method ShadowSense, which they define as a modality between vision and touch, bringing “the high resolution and low cost of vision-sensing to the close-up sensory experience of touch.”
Touch-sensing in social or interactive robots is usually achieved with force sensors or capacitive sensors, says study co-author Guy Hoffman of the Sibley School of Mechanical and Aerospace Engineering at Cornell University. The drawback to his group’s approach has been that, even to achieve coarse spatial resolution, many sensors are needed in a small area.
However, working with non-rigid, inflatable robots, Hoffman and his co-researchers installed a consumer-grade USB camera to which they attached a fisheye lens for a wider field of vision.
“Given that the robot is already hollow, and has a soft and translucent skin, we could do touch interaction by looking at the shadows created by people touching the robot,” says Hoffman. They used deep neural networks to interpret the shadows. “And we were able to do it with very high accuracy,” he says. The robot was able to interpret six different gestures, including one- or two-handed touch, pointing, hugging and punching, with an accuracy of 87.5 to 96 percent, depending on the lighting.
This is not the first time that computer vision has been used for tactile sensing, though the scale and application of ShadowSense is unique. “Photography has been used for touch mainly in robotic grasping,” says Hoffman. By contrast, Hoffman and collaborators wanted to develop a sense that could be “felt” across the whole of the device.
The potential applications for ShadowSense include mobile robot guidance using touch, and interactive screens on soft robots. A third concerns privacy, especially in home-based social robots. “We have another paper currently under review that looks specifically at the ability to detect gestures that are further away [from the robot’s skin],” says Hoffman. This way, users would be able to cover their robot’s camera with a translucent material and still allow it to interpret actions and gestures from shadows. Thus, even though it’s prevented from capturing a high-resolution image of the user or their surrounding environment, using the right kind of training datasets, the robot can continue to monitor some kinds of non-tactile activities.
In its current iteration, Hoffman says, ShadowSense doesn’t do well in low-light conditions. Environmental noise, or shadows from surrounding objects, also interfere with image classification. Relying on one camera also means a single point of failure. “I think if this were to become a commercial product, we would probably [have to] work a little bit better on image detection,” says Hoffman.
As it was, the researchers used transfer learning—reusing a pre-trained deep-learning model in a new problem—for image analysis. “One of the problems with multi-layered neural networks is that you need a lot of training data to make accurate predictions,” says Hoffman. “Obviously, we don’t have millions of examples of people touching a hollow, inflatable robot. But we can use pre-trained networks trained on general images, which we have billions of, and we only retrain the last layers of the network using our own dataset.” Continue reading
Posted in Human Robots
#438798 This AI Thrashes the Hardest Atari Games ...
Learning from rewards seems like the simplest thing. I make coffee, I sip coffee, I’m happy. My brain registers “brewing coffee” as an action that leads to a reward.
That’s the guiding insight behind deep reinforcement learning, a family of algorithms that famously smashed most of Atari’s gaming catalog and triumphed over humans in strategy games like Go. Here, an AI “agent” explores the game, trying out different actions and registering ones that let it win.
Except it’s not that simple. “Brewing coffee” isn’t one action; it’s a series of actions spanning several minutes, where you’re only rewarded at the very end. By just tasting the final product, how do you learn to fine-tune grind coarseness, water to coffee ratio, brewing temperature, and a gazillion other factors that result in the reward—tasty, perk-me-up coffee?
That’s the problem with “sparse rewards,” which are ironically very abundant in our messy, complex world. We don’t immediately get feedback from our actions—no video-game-style dings or points for just grinding coffee beans—yet somehow we’re able to learn and perform an entire sequence of arm and hand movements while half-asleep.
This week, researchers from UberAI and OpenAI teamed up to bestow this talent on AI.
The trick is to encourage AI agents to “return” to a previous step, one that’s promising for a winning solution. The agent then keeps a record of that state, reloads it, and branches out again to intentionally explore other solutions that may have been left behind on the first go-around. Video gamers are likely familiar with this idea: live, die, reload a saved point, try something else, repeat for a perfect run-through.
The new family of algorithms, appropriately dubbed “Go-Explore,” smashed notoriously difficult Atari games like Montezuma’s Revenge that were previously unsolvable by its AI predecessors, while trouncing human performance along the way.
It’s not just games and digital fun. In a computer simulation of a robotic arm, the team found that installing Go-Explore as its “brain” allowed it to solve a challenging series of actions when given very sparse rewards. Because the overarching idea is so simple, the authors say, it can be adapted and expanded to other real-world problems, such as drug design or language learning.
Growing Pains
How do you reward an algorithm?
Rewards are very hard to craft, the authors say. Take the problem of asking a robot to go to a fridge. A sparse reward will only give the robot “happy points” if it reaches its destination, which is similar to asking a baby, with no concept of space and danger, to crawl through a potential minefield of toys and other obstacles towards a fridge.
“In practice, reinforcement learning works very well, if you have very rich feedback, if you can tell, ‘hey, this move is good, that move is bad, this move is good, that move is bad,’” said study author Joost Huinzinga. However, in situations that offer very little feedback, “rewards can intentionally lead to a dead end. Randomly exploring the space just doesn’t cut it.”
The other extreme is providing denser rewards. In the same robot-to-fridge example, you could frequently reward the bot as it goes along its journey, essentially helping “map out” the exact recipe to success. But that’s troubling as well. Over-holding an AI’s hand could result in an extremely rigid robot that ignores new additions to its path—a pet, for example—leading to dangerous situations. It’s a deceptive AI solution that seems effective in a simple environment, but crashes in the real world.
What we need are AI agents that can tackle both problems, the team said.
Intelligent Exploration
The key is to return to the past.
For AI, motivation usually comes from “exploring new or unusual situations,” said Huizinga. It’s efficient, but comes with significant downsides. For one, the AI agent could prematurely stop going back to promising areas because it thinks it had already found a good solution. For another, it could simply forget a previous decision point because of the mechanics of how it probes the next step in a problem.
For a complex task, the end result is an AI that randomly stumbles around towards a solution while ignoring potentially better ones.
“Detaching from a place that was previously visited after collecting a reward doesn’t work in difficult games, because you might leave out important clues,” Huinzinga explained.
Go-Explore solves these problems with a simple principle: first return, then explore. In essence, the algorithm saves different approaches it previously tried and loads promising save points—once more likely to lead to victory—to explore further.
Digging a bit deeper, the AI stores screen caps from a game. It then analyzes saved points and groups images that look alike as a potential promising “save point” to return to. Rinse and repeat. The AI tries to maximize its final score in the game, and updates its save points when it achieves a new record score. Because Atari doesn’t usually allow people to revisit any random point, the team used an emulator, which is a kind of software that mimics the Atari system but with custom abilities such as saving and reloading at any time.
The trick worked like magic. When pitted against 55 Atari games in the OpenAI gym, now commonly used to benchmark reinforcement learning algorithms, Go-Explore knocked out state-of-the-art AI competitors over 85 percent of the time.
It also crushed games previously unbeatable by AI. Montezuma’s Revenge, for example, requires you to move Pedro, the blocky protagonist, through a labyrinth of underground temples while evading obstacles such as traps and enemies and gathering jewels. One bad jump could derail the path to the next level. It’s a perfect example of sparse rewards: you need a series of good actions to get to the reward—advancing onward.
Go-Explore didn’t just beat all levels of the game, a first for AI. It also scored higher than any previous record for reinforcement learning algorithms at lower levels while toppling the human world record.
Outside a gaming environment, Go-Explore was also able to boost the performance of a simulated robot arm. While it’s easy for humans to follow high-level guidance like “put the cup on this shelf in a cupboard,” robots often need explicit training—from grasping the cup to recognizing a cupboard, moving towards it while avoiding obstacles, and learning motions to not smash the cup when putting it down.
Here, similar to the real world, the digital robot arm was only rewarded when it placed the cup onto the correct shelf, out of four possible shelves. When pitted against another algorithm, Go-Explore quickly figured out the movements needed to place the cup, while its competitor struggled with even reliably picking the cup up.
Combining Forces
By itself, the “first return, then explore” idea behind Go-Explore is already powerful. The team thinks it can do even better.
One idea is to change the mechanics of save points. Rather than reloading saved states through the emulator, it’s possible to train a neural network to do the same, without needing to relaunch a saved state. It’s a potential way to make the AI even smarter, the team said, because it can “learn” to overcome one obstacle once, instead of solving the same problem again and again. The downside? It’s much more computationally intensive.
Another idea is to combine Go-Explore with an alternative form of learning, called “imitation learning.” Here, an AI observes human behavior and mimics it through a series of actions. Combined with Go-Explore, said study author Adrien Ecoffet, this could make more robust robots capable of handling all the complexity and messiness in the real world.
To the team, the implications go far beyond Go-Explore. The concept of “first return, then explore” seems to be especially powerful, suggesting “it may be a fundamental feature of learning in general.” The team said, “Harnessing these insights…may be essential…to create generally intelligent agents.”
Image Credit: Adrien Ecoffet, Joost Huizinga, Joel Lehman, Kenneth O. Stanley, and Jeff Clune Continue reading
Posted in Human Robots