Transfer learning, the ability to use knowledge previously gained from one context in another, could teach cheap robots to perform as well as expensive ones....
The context: One of the hardest challenges currently facing robotics is getting the robot to operate smoothly outside the lab. In a research setting, it’s feasible to equip the robot with expensive sensors and provide it an ideal environment to learn navigation. But in the real world, using the same sensors would prove costly and unfriendly for consumers. Plus, it is messy and imperfect.
The proposal: Researchers at Vrije Universiteit turned to a type of machine learning known as transfer learning to see if they could solve the problem. Transfer learning is the process of taking what an algorithm has learned in one context and applying it in another. It could be used to adapt an algorithm that controls a robot in the lab so it can control a robot in the real world. That means the robot could first train with the advantage of better sensors and a better environment, and then exploit what it learned even when it only has cheap sensors and a poor environment.
The results: To test this idea, the researchers created a robot in a simulated environment that it navigated first with the aid of eight proximity sensors, and then with a single camera. They found that when the robot-controlling algorithm used transfer learning to make decisions—with camera access only—it learned to navigate around the room much faster than when it used no transfer learning at all. It was also much faster than when it used transfer learning during training rather than decision-making.
To have more stories like this delivered directly to your inbox, sign up for our Webby-nominated AI newsletter The Algorithm. It's free.
The apps, which could be used to spy on other people’s devices, had been downloaded a total of 130,000 times....
The news: Researchers at cybersecurity firm Avast discovered seven stalkerware apps on the Play Store this week. The most-installed apps were Spy Tracker and SMS Tracker, with more than 50,000 downloads each. The researchers reported them to Google, and they have all been removed.
How they worked: They all required access to the smartphone the perpetrator wanted to spy on. The snoop could download them from the Play Store, install them on the target device, and then enter their own email address and a password to get access to the spying app from their own device.
The apps were all able to track the surveilled person’s location, contacts, SMS messages, and call history. Once installed, there’s no app icon, so the targeted person would have no idea that the stalkerware app was on the phone. Some stalkerware apps don’t even require access to the device, and can be sent disguised as a picture message.
Who are the targets? Although these apps are often used by abusive partners to spy on their victims, they are also sometimes used to track children or even employees.
The scale of the problem: There have been very few studies on stalkerware, so it’s hard to know how big this problem truly is. However, technology is playing a growing role in abusive relationships. Domestic-violence charity Refuge estimates that around 95% of its cases involve some form of technology-based abuse.
Big tech firms are yet to fully face up to, and act upon, the use of their products for this purpose. Google’s action this week is positive, but it would be good to see the companies proactively rooting out this problem, rather than just relying on being prodded by outside experts.
Read next: Head over to this story if you want to read about one woman’s experience of stalkerware, or want advice if you fear you might be a victim of it too.
Sign up here for our daily newsletter The Download to get your dose of the latest must-read news from the world of emerging tech.
Innovators Under 35 | 2019
It’s part of our ethos that technology can and should be a force for good. In these profiles you’ll find people employing innovative methods to treat disease, to fight online harassment, and to create the next big battery breakthrough.
Rediet Abebe
She uses algorithms and AI to fight socioeconomic inequality
Tim Ellis
He developed a massive 3D metal printer—for building an entire rocket
Ritu Raman
She’s developed inchworm-size robots made partly of biological tissue and muscle
John Porter
His innovations could make all kinds of products more accessible to people with disabilities
Kimberly Stachenfeld
She used reinforcement learning to better understand problem solving in both the human brain and AI systems
Advertisement