Tag Archives: dog
#437466 How Future AI Could Recognize a Kangaroo ...
AI is continuously taking on new challenges, from detecting deepfakes (which, incidentally, are also made using AI) to winning at poker to giving synthetic biology experiments a boost. These impressive feats result partly from the huge datasets the systems are trained on. That training is costly and time-consuming, and it yields AIs that can really only do one thing well.
For example, to train an AI to differentiate between a picture of a dog and one of a cat, it’s fed thousands—if not millions—of labeled images of dogs and cats. A child, on the other hand, can see a dog or cat just once or twice and remember which is which. How can we make AIs learn more like children do?
A team at the University of Waterloo in Ontario has an answer: change the way AIs are trained.
Here’s the thing about the datasets normally used to train AI—besides being huge, they’re highly specific. A picture of a dog can only be a picture of a dog, right? But what about a really small dog with a long-ish tail? That sort of dog, while still being a dog, looks more like a cat than, say, a fully-grown Golden Retriever.
It’s this concept that the Waterloo team’s methodology is based on. They described their work in a paper published on the pre-print (or non-peer-reviewed) server arXiv last month. Teaching an AI system to identify a new class of objects using just one example is what they call “one-shot learning.” But they take it a step further, focusing on “less than one shot learning,” or LO-shot learning for short.
LO-shot learning consists of a system learning to classify various categories based on a number of examples that’s smaller than the number of categories. That’s not the most straightforward concept to wrap your head around, so let’s go back to the dogs and cats example. Say you want to teach an AI to identify dogs, cats, and kangaroos. How could that possibly be done without several clear examples of each animal?
The key, the Waterloo team says, is in what they call soft labels. Unlike hard labels, which label a data point as belonging to one specific class, soft labels tease out the relationship or degree of similarity between that data point and multiple classes. In the case of an AI trained on only dogs and cats, a third class of objects, say, kangaroos, might be described as 60 percent like a dog and 40 percent like a cat (I know—kangaroos probably aren’t the best animal to have thrown in as a third category).
“Soft labels can be used to represent training sets using fewer prototypes than there are classes, achieving large increases in sample efficiency over regular (hard-label) prototypes,” the paper says. Translation? Tell an AI a kangaroo is some fraction cat and some fraction dog—both of which it’s seen and knows well—and it’ll be able to identify a kangaroo without ever having seen one.
If the soft labels are nuanced enough, you could theoretically teach an AI to identify a large number of categories based on a much smaller number of training examples.
The paper’s authors use a simple machine learning algorithm called k-nearest neighbors (kNN) to explore this idea more in depth. The algorithm operates under the assumption that similar things are most likely to exist near each other; if you go to a dog park, there will be lots of dogs but no cats or kangaroos. Go to the Australian grasslands and there’ll be kangaroos but no cats or dogs. And so on.
To train a kNN algorithm to differentiate between categories, you choose specific features to represent each category (i.e. for animals you could use weight or size as a feature). With one feature on the x-axis and the other on the y-axis, the algorithm creates a graph where data points that are similar to each other are clustered near each other. A line down the center divides the categories, and it’s pretty straightforward for the algorithm to discern which side of the line new data points should fall on.
The Waterloo team kept it simple and used plots of color on a 2D graph. Using the colors and their locations on the graphs, the team created synthetic data sets and accompanying soft labels. One of the more simplistic graphs is pictured below, along with soft labels in the form of pie charts.
Image Credit: Ilia Sucholutsky & Matthias Schonlau
When the team had the algorithm plot the boundary lines of the different colors based on these soft labels, it was able to split the plot up into more colors than the number of data points it was given in the soft labels.
While the results are encouraging, the team acknowledges that they’re just the first step, and there’s much more exploration of this concept yet to be done. The kNN algorithm is one of the least complex models out there; what might happen when LO-shot learning is applied to a far more complex algorithm? Also, to apply it, you still need to distill a larger dataset down into soft labels.
One idea the team is already working on is having other algorithms generate the soft labels for the algorithm that’s going to be trained using LO-shot; manually designing soft labels won’t always be as easy as splitting up some pie charts into different colors.
LO-shot’s potential for reducing the amount of training data needed to yield working AI systems is promising. Besides reducing the cost and the time required to train new models, the method could also make AI more accessible to industries, companies, or individuals who don’t have access to large datasets—an important step for democratization of AI.
Image Credit: pen_ash from Pixabay Continue reading
Posted in Human Robots
#437276 Cars Will Soon Be Able to Sense and ...
Imagine you’re on your daily commute to work, driving along a crowded highway while trying to resist looking at your phone. You’re already a little stressed out because you didn’t sleep well, woke up late, and have an important meeting in a couple hours, but you just don’t feel like your best self.
Suddenly another car cuts you off, coming way too close to your front bumper as it changes lanes. Your already-simmering emotions leap into overdrive, and you lay on the horn and shout curses no one can hear.
Except someone—or, rather, something—can hear: your car. Hearing your angry words, aggressive tone, and raised voice, and seeing your furrowed brow, the onboard computer goes into “soothe” mode, as it’s been programmed to do when it detects that you’re angry. It plays relaxing music at just the right volume, releases a puff of light lavender-scented essential oil, and maybe even says some meditative quotes to calm you down.
What do you think—creepy? Helpful? Awesome? Weird? Would you actually calm down, or get even more angry that a car is telling you what to do?
Scenarios like this (maybe without the lavender oil part) may not be imaginary for much longer, especially if companies working to integrate emotion-reading artificial intelligence into new cars have their way. And it wouldn’t just be a matter of your car soothing you when you’re upset—depending what sort of regulations are enacted, the car’s sensors, camera, and microphone could collect all kinds of data about you and sell it to third parties.
Computers and Feelings
Just as AI systems can be trained to tell the difference between a picture of a dog and one of a cat, they can learn to differentiate between an angry tone of voice or facial expression and a happy one. In fact, there’s a whole branch of machine intelligence devoted to creating systems that can recognize and react to human emotions; it’s called affective computing.
Emotion-reading AIs learn what different emotions look and sound like from large sets of labeled data; “smile = happy,” “tears = sad,” “shouting = angry,” and so on. The most sophisticated systems can likely even pick up on the micro-expressions that flash across our faces before we consciously have a chance to control them, as detailed by Daniel Goleman in his groundbreaking book Emotional Intelligence.
Affective computing company Affectiva, a spinoff from MIT Media Lab, says its algorithms are trained on 5,313,751 face videos (videos of people’s faces as they do an activity, have a conversation, or react to stimuli) representing about 2 billion facial frames. Fascinatingly, Affectiva claims its software can even account for cultural differences in emotional expression (for example, it’s more normalized in Western cultures to be very emotionally expressive, whereas Asian cultures tend to favor stoicism and politeness), as well as gender differences.
But Why?
As reported in Motherboard, companies like Affectiva, Cerence, Xperi, and Eyeris have plans in the works to partner with automakers and install emotion-reading AI systems in new cars. Regulations passed last year in Europe and a bill just introduced this month in the US senate are helping make the idea of “driver monitoring” less weird, mainly by emphasizing the safety benefits of preemptive warning systems for tired or distracted drivers (remember that part in the beginning about sneaking glances at your phone? Yeah, that).
Drowsiness and distraction can’t really be called emotions, though—so why are they being lumped under an umbrella that has a lot of other implications, including what many may consider an eerily Big Brother-esque violation of privacy?
Our emotions, in fact, are among the most private things about us, since we are the only ones who know their true nature. We’ve developed the ability to hide and disguise our emotions, and this can be a useful skill at work, in relationships, and in scenarios that require negotiation or putting on a game face.
And I don’t know about you, but I’ve had more than one good cry in my car. It’s kind of the perfect place for it; private, secluded, soundproof.
Putting systems into cars that can recognize and collect data about our emotions under the guise of preventing accidents due to the state of mind of being distracted or the physical state of being sleepy, then, seems a bit like a bait and switch.
A Highway to Privacy Invasion?
European regulations will help keep driver data from being used for any purpose other than ensuring a safer ride. But the US is lagging behind on the privacy front, with car companies largely free from any enforceable laws that would keep them from using driver data as they please.
Affectiva lists the following as use cases for occupant monitoring in cars: personalizing content recommendations, providing alternate route recommendations, adapting environmental conditions like lighting and heating, and understanding user frustration with virtual assistants and designing those assistants to be emotion-aware so that they’re less frustrating.
Our phones already do the first two (though, granted, we’re not supposed to look at them while we drive—but most cars now let you use bluetooth to display your phone’s content on the dashboard), and the third is simply a matter of reaching a hand out to turn a dial or press a button. The last seems like a solution for a problem that wouldn’t exist without said… solution.
Despite how unnecessary and unsettling it may seem, though, emotion-reading AI isn’t going away, in cars or other products and services where it might provide value.
Besides automotive AI, Affectiva also makes software for clients in the advertising space. With consent, the built-in camera on users’ laptops records them while they watch ads, gauging their emotional response, what kind of marketing is most likely to engage them, and how likely they are to buy a given product. Emotion-recognition tech is also being used or considered for use in mental health applications, call centers, fraud monitoring, and education, among others.
In a 2015 TED talk, Affectiva co-founder Rana El-Kaliouby told her audience that we’re living in a world increasingly devoid of emotion, and her goal was to bring emotions back into our digital experiences. Soon they’ll be in our cars, too; whether the benefits will outweigh the costs remains to be seen.
Image Credit: Free-Photos from Pixabay Continue reading
Posted in Human Robots
#437269 DeepMind’s Newest AI Programs Itself ...
When Deep Blue defeated world chess champion Garry Kasparov in 1997, it may have seemed artificial intelligence had finally arrived. A computer had just taken down one of the top chess players of all time. But it wasn’t to be.
Though Deep Blue was meticulously programmed top-to-bottom to play chess, the approach was too labor-intensive, too dependent on clear rules and bounded possibilities to succeed at more complex games, let alone in the real world. The next revolution would take a decade and a half, when vastly more computing power and data revived machine learning, an old idea in artificial intelligence just waiting for the world to catch up.
Today, machine learning dominates, mostly by way of a family of algorithms called deep learning, while symbolic AI, the dominant approach in Deep Blue’s day, has faded into the background.
Key to deep learning’s success is the fact the algorithms basically write themselves. Given some high-level programming and a dataset, they learn from experience. No engineer anticipates every possibility in code. The algorithms just figure it.
Now, Alphabet’s DeepMind is taking this automation further by developing deep learning algorithms that can handle programming tasks which have been, to date, the sole domain of the world’s top computer scientists (and take them years to write).
In a paper recently published on the pre-print server arXiv, a database for research papers that haven’t been peer reviewed yet, the DeepMind team described a new deep reinforcement learning algorithm that was able to discover its own value function—a critical programming rule in deep reinforcement learning—from scratch.
Surprisingly, the algorithm was also effective beyond the simple environments it trained in, going on to play Atari games—a different, more complicated task—at a level that was, at times, competitive with human-designed algorithms and achieving superhuman levels of play in 14 games.
DeepMind says the approach could accelerate the development of reinforcement learning algorithms and even lead to a shift in focus, where instead of spending years writing the algorithms themselves, researchers work to perfect the environments in which they train.
Pavlov’s Digital Dog
First, a little background.
Three main deep learning approaches are supervised, unsupervised, and reinforcement learning.
The first two consume huge amounts of data (like images or articles), look for patterns in the data, and use those patterns to inform actions (like identifying an image of a cat). To us, this is a pretty alien way to learn about the world. Not only would it be mind-numbingly dull to review millions of cat images, it’d take us years or more to do what these programs do in hours or days. And of course, we can learn what a cat looks like from just a few examples. So why bother?
While supervised and unsupervised deep learning emphasize the machine in machine learning, reinforcement learning is a bit more biological. It actually is the way we learn. Confronted with several possible actions, we predict which will be most rewarding based on experience—weighing the pleasure of eating a chocolate chip cookie against avoiding a cavity and trip to the dentist.
In deep reinforcement learning, algorithms go through a similar process as they take action. In the Atari game Breakout, for instance, a player guides a paddle to bounce a ball at a ceiling of bricks, trying to break as many as possible. When playing Breakout, should an algorithm move the paddle left or right? To decide, it runs a projection—this is the value function—of which direction will maximize the total points, or rewards, it can earn.
Move by move, game by game, an algorithm combines experience and value function to learn which actions bring greater rewards and improves its play, until eventually, it becomes an uncanny Breakout player.
Learning to Learn (Very Meta)
So, a key to deep reinforcement learning is developing a good value function. And that’s difficult. According to the DeepMind team, it takes years of manual research to write the rules guiding algorithmic actions—which is why automating the process is so alluring. Their new Learned Policy Gradient (LPG) algorithm makes solid progress in that direction.
LPG trained in a number of toy environments. Most of these were “gridworlds”—literally two-dimensional grids with objects in some squares. The AI moves square to square and earns points or punishments as it encounters objects. The grids vary in size, and the distribution of objects is either set or random. The training environments offer opportunities to learn fundamental lessons for reinforcement learning algorithms.
Only in LPG’s case, it had no value function to guide that learning.
Instead, LPG has what DeepMind calls a “meta-learner.” You might think of this as an algorithm within an algorithm that, by interacting with its environment, discovers both “what to predict,” thereby forming its version of a value function, and “how to learn from it,” applying its newly discovered value function to each decision it makes in the future.
Prior work in the area has had some success, but according to DeepMind, LPG is the first algorithm to discover reinforcement learning rules from scratch and to generalize beyond training. The latter was particularly surprising because Atari games are so different from the simple worlds LPG trained in—that is, it had never seen anything like an Atari game.
Time to Hand Over the Reins? Not Just Yet
LPG is still behind advanced human-designed algorithms, the researchers said. But it outperformed a human-designed benchmark in training and even some Atari games, which suggests it isn’t strictly worse, just that it specializes in some environments.
This is where there’s room for improvement and more research.
The more environments LPG saw, the more it could successfully generalize. Intriguingly, the researchers speculate that with enough well-designed training environments, the approach might yield a general-purpose reinforcement learning algorithm.
At the least, though, they say further automation of algorithm discovery—that is, algorithms learning to learn—will accelerate the field. In the near term, it can help researchers more quickly develop hand-designed algorithms. Further out, as self-discovered algorithms like LPG improve, engineers may shift from manually developing the algorithms themselves to building the environments where they learn.
Deep learning long ago left Deep Blue in the dust at games. Perhaps algorithms learning to learn will be a winning strategy in the real world too.
Image credit: Mike Szczepanski / Unsplash Continue reading
Posted in Human Robots