Tag Archives: blocks
#435707 AI Agents Startle Researchers With ...
After 25 million games, the AI agents playing hide-and-seek with each other had mastered four basic game strategies. The researchers expected that part.
After a total of 380 million games, the AI players developed strategies that the researchers didn’t know were possible in the game environment—which the researchers had themselves created. That was the part that surprised the team at OpenAI, a research company based in San Francisco.
The AI players learned everything via a machine learning technique known as reinforcement learning. In this learning method, AI agents start out by taking random actions. Sometimes those random actions produce desired results, which earn them rewards. Via trial-and-error on a massive scale, they can learn sophisticated strategies.
In the context of games, this process can be abetted by having the AI play against another version of itself, ensuring that the opponents will be evenly matched. It also locks the AI into a process of one-upmanship, where any new strategy that emerges forces the opponent to search for a countermeasure. Over time, this “self-play” amounted to what the researchers call an “auto-curriculum.”
According to OpenAI researcher Igor Mordatch, this experiment shows that self-play “is enough for the agents to learn surprising behaviors on their own—it’s like children playing with each other.”
Reinforcement is a hot field of AI research right now. OpenAI’s researchers used the technique when they trained a team of bots to play the video game Dota 2, which squashed a world-champion human team last April. The Alphabet subsidiary DeepMind has used it to triumph in the ancient board game Go and the video game StarCraft.
Aniruddha Kembhavi, a researcher at the Allen Institute for Artificial Intelligence (AI2) in Seattle, says games such as hide-and-seek offer a good way for AI agents to learn “foundational skills.” He worked on a team that taught their AllenAI to play Pictionary with humans, viewing the gameplay as a way for the AI to work on common sense reasoning and communication. “We are, however, quite far away from being able to translate these preliminary findings in highly simplified environments into the real world,” says Kembhavi.
Illustration: OpenAI
AI agents construct a fort during a hide-and-seek game developed by OpenAI.
In OpenAI’s game of hide-and-seek, both the hiders and the seekers received a reward only if they won the game, leaving the AI players to develop their own strategies. Within a simple 3D environment containing walls, blocks, and ramps, the players first learned to run around and chase each other (strategy 1). The hiders next learned to move the blocks around to build forts (2), and then the seekers learned to move the ramps (3), enabling them to jump inside the forts. Then the hiders learned to move all the ramps into their forts before the seekers could use them (4).
The two strategies that surprised the researchers came next. First the seekers learned that they could jump onto a box and “surf” it over to a fort (5), allowing them to jump in—a maneuver that the researchers hadn’t realized was physically possible in the game environment. So as a final countermeasure, the hiders learned to lock all the boxes into place (6) so they weren’t available for use as surfboards.
Illustration: OpenAI
An AI agent uses a nearby box to surf its way into a competitor’s fort.
In this circumstance, having AI agents behave in an unexpected way wasn’t a problem: They found different paths to their rewards, but didn’t cause any trouble. However, you can imagine situations in which the outcome would be rather serious. Robots acting in the real world could do real damage. And then there’s Nick Bostrom’s famous example of a paper clip factory run by an AI, whose goal is to make as many paper clips as possible. As Bostrom told IEEE Spectrum back in 2014, the AI might realize that “human bodies consist of atoms, and those atoms could be used to make some very nice paper clips.”
Bowen Baker, another member of the OpenAI research team, notes that it’s hard to predict all the ways an AI agent will act inside an environment—even a simple one. “Building these environments is hard,” he says. “The agents will come up with these unexpected behaviors, which will be a safety problem down the road when you put them in more complex environments.”
AI researcher Katja Hofmann at Microsoft Research Cambridge, in England, has seen a lot of gameplay by AI agents: She started a competition that uses Minecraft as the playing field. She says the emergent behavior seen in this game, and in prior experiments by other researchers, shows that games can be a useful for studies of safe and responsible AI.
“I find demonstrations like this, in games and game-like settings, a great way to explore the capabilities and limitations of existing approaches in a safe environment,” says Hofmann. “Results like these will help us develop a better understanding on how to validate and debug reinforcement learning systems–a crucial step on the path towards real-world applications.”
Baker says there’s also a hopeful takeaway from the surprises in the hide-and-seek experiment. “If you put these agents into a rich enough environment they will find strategies that we never knew were possible,” he says. “Maybe they can solve problems that we can’t imagine solutions to.” Continue reading
Posted in Human Robots
#435664 Swarm Robots Mimic Ant Jaws to Flip and ...
Small robots are appealing because they’re simple, cheap, and it’s easy to make a lot of them. Unfortunately, being simple and cheap means that each robot individually can’t do a whole lot. To make up for this, you can do what insects do—leverage that simplicity and low-cost to just make a huge swarm of simple robots, and together, they can cooperate to carry out relatively complex tasks.
Using insects as an example does set a bit of an unfair expectation for the poor robots, since insects are (let’s be honest) generally smarter and much more versatile than a robot on their scale could ever hope to be. Most robots with insect-like capabilities (like DASH and its family) are really too big and complex to be turned into swarms, because to make a vast amount of small robots, things like motors aren’t going to work because they’re too expensive.
The question, then, is to how to make a swarm of inexpensive small robots with insect-like mobility that don’t need motors to get around, and Jamie Paik’s Reconfigurable Robotics Lab at EPFL has an answer, inspired by trap-jaw ants.
Let’s talk about trap-jaw ants for just a second, because they’re insane. You can read this 2006 paper about them if you’re particularly interested in insane ants (and who isn’t!), but if you just want to hear the insane bit, it’s that trap-jaw ants can fire themselves into the air by biting the ground (!). In just 0.06 millisecond, their half-millimeter long mandibles can close at a top speed of 64 meters per second, which works out to an acceleration of about 100,000 g’s. Biting the ground causes the ant’s head to snap back with a force of 300 times the body weight of the ant itself, which launches the ant upwards. The ants can fly 8 centimeters vertically, and up to 15 cm horizontally—this is a lot, for an ant that’s just a few millimeters long.
Trap-jaw ants can fire themselves into the air by biting the ground, causing the ant’s head to snap back with a force of 300 times the body weight of the ant itself
EPFL’s robots, called Tribots, look nothing at all like trap-jaw ants, which personally I am fine with. They’re about 5 cm tall, weighing 10 grams each, and can be built on a flat sheet, and then folded into a tripod shape, origami-style. Or maybe it’s kirigami, because there’s some cutting involved. The Tribots are fully autonomous, meaning they have onboard power and control, including proximity sensors that allow them to detect objects and avoid them.
Photo: Marc Delachaux/EPFL
EPFL researchers Zhenishbek Zhakypov and Jamie Paik.
Avoiding objects is where the trap-jaw ants come in. Using two different shape-memory actuators (a spring and a latch, similar to how the ant’s jaw works), the Tribots can move around using a bunch of different techniques that can adapt to the terrain that they’re on, including:
Vertical jumping for height
Horizontal jumping for distance
Somersault jumping to clear obstacles
Walking on textured terrain with short hops (called “flic-flac” walking)
Crawling on flat surfaces
Here’s the robot in action:
Tribot’s maximum vertical jump is 14 cm (2.5 times its height), and horizontally it can jump about 23 cm (almost 4 times its length). Tribot is actually quite efficient in these movements, with a cost of transport much lower than similarly-sized robots, on par with insects themselves.
Working together, small groups of Tribots can complete tasks that a single robot couldn’t do alone. One example is pushing a heavy object a set distance. It turns out that you need five Tribots for this task—a leader robot, two worker robots, a monitor robot to measure the distance that the object has been pushed, and then a messenger robot to relay communications around the obstacle.
Image: EPFL
Five Tribots collaborate to move an object to a desired position, using coordination between a leader, two workers, a monitor, and a messenger robot. The leader orders the two worker robots to push the object while the monitor measures the relative position of the object. As the object blocks the two-way link between the leader and the monitor, the messenger maintains the communication link.
The researchers acknowledge that the current version of the hardware is limited in pretty much every way (mobility, sensing, and computation), but it does a reasonable job of demonstrating what’s possible with the concept. The plan going forward is to automate fabrication in order to “enable on-demand, ’push-button-manufactured’” robots.
“Designing minimal and scalable insect-inspired multi-locomotion millirobots,” by Zhenishbek Zhakypov, Kazuaki Mori, Koh Hosoda, and Jamie Paik from EPFL and Osaka University, is published in the current issue of Nature.
[ RRL ] via [ EPFL ] Continue reading
Posted in Human Robots
#435474 Watch China’s New Hybrid AI Chip Power ...
When I lived in Beijing back in the 90s, a man walking his bike was nothing to look at. But today, I did a serious double-take at a video of a bike walking his man.
No kidding.
The bike itself looks overloaded but otherwise completely normal. Underneath its simplicity, however, is a hybrid computer chip that combines brain-inspired circuits with machine learning processes into a computing behemoth. Thanks to its smart chip, the bike self-balances as it gingerly rolls down a paved track before smoothly gaining speed into a jogging pace while navigating dexterously around obstacles. It can even respond to simple voice commands such as “speed up,” “left,” or “straight.”
Far from a circus trick, the bike is a real-world demo of the AI community’s latest attempt at fashioning specialized hardware to keep up with the challenges of machine learning algorithms. The Tianjic (天机*) chip isn’t just your standard neuromorphic chip. Rather, it has the architecture of a brain-like chip, but can also run deep learning algorithms—a match made in heaven that basically mashes together neuro-inspired hardware and software.
The study shows that China is readily nipping at the heels of Google, Facebook, NVIDIA, and other tech behemoths investing in developing new AI chip designs—hell, with billions in government investment it may have already had a head start. A sweeping AI plan from 2017 looks to catch up with the US on AI technology and application by 2020. By 2030, China’s aiming to be the global leader—and a champion for building general AI that matches humans in intellectual competence.
The country’s ambition is reflected in the team’s parting words.
“Our study is expected to stimulate AGI [artificial general intelligence] development by paving the way to more generalized hardware platforms,” said the authors, led by Dr. Luping Shi at Tsinghua University.
A Hardware Conundrum
Shi’s autonomous bike isn’t the first robotic two-wheeler. Back in 2015, the famed research nonprofit SRI International in Menlo Park, California teamed up with Yamaha to engineer MOTOBOT, a humanoid robot capable of driving a motorcycle. Powered by state-of-the-art robotic hardware and machine learning, MOTOBOT eventually raced MotoGPTM world champion Valentino Rossi in a nail-biting match-off.
However, the technological core of MOTOBOT and Shi’s bike vastly differ, and that difference reflects two pathways towards more powerful AI. One, exemplified by MOTOBOT, is software—developing brain-like algorithms with increasingly efficient architecture, efficacy, and speed. That sounds great, but deep neural nets demand so many computational resources that general-purpose chips can’t keep up.
As Shi told China Science Daily: “CPUs and other chips are driven by miniaturization technologies based on physics. Transistors might shrink to nanoscale-level in 10, 20 years. But what then?” As more transistors are squeezed onto these chips, efficient cooling becomes a limiting factor in computational speed. Tax them too much, and they melt.
For AI processes to continue, we need better hardware. An increasingly popular idea is to build neuromorphic chips, which resemble the brain from the ground up. IBM’s TrueNorth, for example, contains a massively parallel architecture nothing like the traditional Von Neumann structure of classic CPUs and GPUs. Similar to biological brains, TrueNorth’s memory is stored within “synapses” between physical “neurons” etched onto the chip, which dramatically cuts down on energy consumption.
But even these chips are limited. Because computation is tethered to hardware architecture, most chips resemble just one specific type of brain-inspired network called spiking neural networks (SNNs). Without doubt, neuromorphic chips are highly efficient setups with dynamics similar to biological networks. They also don’t play nicely with deep learning and other software-based AI.
Brain-AI Hybrid Core
Shi’s new Tianjic chip brought the two incompatibilities together onto a single piece of brainy hardware.
First was to bridge the deep learning and SNN divide. The two have very different computation philosophies and memory organizations, the team said. The biggest difference, however, is that artificial neural networks transform multidimensional data—image pixels, for example—into a single, continuous, multi-bit 0 and 1 stream. In contrast, neurons in SNNs activate using something called “binary spikes” that code for specific activation events in time.
Confused? Yeah, it’s hard to wrap my head around it too. That’s because SNNs act very similarly to our neural networks and nothing like computers. A particular neuron needs to generate an electrical signal (a “spike”) large enough to transfer down to the next one; little blips in signals don’t count. The way they transmit data also heavily depends on how they’re connected, or the network topology. The takeaway: SNNs work pretty differently than deep learning.
Shi’s team first recreated this firing quirk in the language of computers—0s and 1s—so that the coding mechanism would become compatible with deep learning algorithms. They then carefully aligned the step-by-step building blocks of the two models, which allowed them to tease out similarities into a common ground to further build on. “On the basis of this unified abstraction, we built a cross-paradigm neuron scheme,” they said.
In general, the design allowed both computational approaches to share the synapses, where neurons connect and store data, and the dendrites, the outgoing branches of the neurons. In contrast, the neuron body, where signals integrate, was left reconfigurable for each type of computation, as were the input branches. Each building block was combined into a single unified functional core (FCore), which acts like a deep learning/SNN converter depending on its specific setup. Translation: the chip can do both types of previously incompatible computation.
The Chip
Using nanoscale fabrication, the team arranged 156 FCores, containing roughly 40,000 neurons and 10 million synapses, onto a chip less than a fifth of an inch in length and width. Initial tests showcased the chip’s versatility, in that it can run both SNNs and deep learning algorithms such as the popular convolutional neural network (CNNs) often used in machine vision.
Compared to IBM TrueNorth, the density of Tianjic’s cores increased by 20 percent, speeding up performance ten times and increasing bandwidth at least 100-fold, the team said. When pitted against GPUs, the current hardware darling of machine learning, the chip increased processing throughput up to 100 times, while using just a sliver (1/10,000) of energy.
Although these stats are great, real-life performance is even better as a demo. Here’s where the authors gave their Tianjic brain a body. The team combined one chip with multiple specialized networks to process vision, balance, voice commands, and decision-making in real time. Object detection and target tracking, for example, relied on a deep neural net CNN, whereas voice commands and balance data were recognized using an SNN. The inputs were then integrated inside a neural state machine, which churned out decisions to downstream output modules—for example, controlling the handle bar to turn left.
Thanks to the chip’s brain-like architecture and bilingual ability, Tianjic “allowed all of the neural network models to operate in parallel and realized seamless communication across the models,” the team said. The result is an autonomous bike that rolls after its human, balances across speed bumps, avoids crashing into roadblocks, and answers to voice commands.
General AI?
“It’s a wonderful demonstration and quite impressive,” said the editorial team at Nature, which published the study on its cover last week.
However, they cautioned, when comparing Tianjic with state-of-the-art chips designed for a single problem toe-to-toe on that particular problem, Tianjic falls behind. But building these jack-of-all-trades hybrid chips is definitely worth the effort. Compared to today’s limited AI, what people really want is artificial general intelligence, which will require new architectures that aren’t designed to solve one particular problem.
Until people start to explore, innovate, and play around with different designs, it’s not clear how we can further progress in the pursuit of general AI. A self-driving bike might not be much to look at, but its hybrid brain is a pretty neat place to start.
*The name, in Chinese, means “heavenly machine,” “unknowable mystery of nature,” or “confidentiality.” Go figure.
Image Credit: Alexander Ryabintsev / Shutterstock.com Continue reading
Posted in Human Robots