We’ve seen a lot of innovation and progress in facial recognition recently. Last month, for example, Apple acquired RealFace, whose facial recognition technology can be used to authenticate users (think selfie passwords). But ‘with great facial recognition technology comes great responsibility.’
Now, facial recognition is being discussed in the news in connection with police body cams. The technology already exists which allows police officers to recognise people in real time, but people debate whether or not it is ethical. Some say that it is problematical since it puts people ‘under perpetual surveillance and suspicion.’
Police body cameras are widely seen as a way to improve law enforcement’s transparency with the public. But when mixed with police use of facial-recognition tools, the prospect of continual surveillance comes with big risks to privacy.
Facial-recognition technology combined with policy body cameras could ‘redefine the nature of public spaces,’ Alvaro Bedoya, executive director of the Georgetown Law Center on Privacy & Technology, told the US House Oversight Committee at a hearing on March 22. It’s not a distant reality and it threatens civil liberties, he warned.
Researchers at OpenAI have discovered that evolution strategies (ES), an optimization technique that’s been known for decades, rivals the performance of standard reinforcement learning (RL) techniques on modern RL benchmarks (e.g. Atari/MuJoCo), while overcoming many of RL’s inconveniences.
In particular, ES is simpler to implement (there is no need for backpropagation), it is easier to scale in a distributed setting, it does not suffer in settings with sparse rewards, and has fewer hyperparameters. This outcome is surprising because ES resembles simple hill-climbing in a high-dimensional space based only on finite differences along a few random directions at each step.
Adelyn Zhou, from TOPBOTS, asked 21 bot experts, CEOs, and other executives to share their predictions for how bots will continue to evolve in the coming year.
Andy Mauro, CEO of Automat, for example said that “in 2017 brands will realise that Conversational Marketing is a better way to learn about and build relationships with their customer than today’s digital marketing which monitors their customers with cookies, pixels, search and social data.”
In the world of of Artificial Intelligence there are all sorts of acronyms which are thrown around: AI (Artificial Intelligence), ML (Machine Learning), and NLP (Natural Language Processing). But what do they mean and how do they relate to one another? Astro put together a handy blogpost which explains the terms in a clear and concise manner. Check it out here.
A British teenager has contacted scientists at Nasa to point out an error in a set of their own data. A-level student Miles Soloman found that radiation sensors on the International Space Station (ISS) were recording false data.
The 17-year-old from Tapton school in Sheffield said it was “pretty cool” to email the space agency. The correction was said to be “appreciated” by Nasa, which invited him to help analyse the problem.
Last week, we reported on how Facebook is using machine learning to detect fake news. Recently, Co.Design spoke with social media’s biggest players to find out how the battle against mistruth is progressing. It’s a long read, but a worthwhile one, since it goes into a lot of detail about the issue and how it is being combatted.
Some, such as Apple, are tackling the problem by only running specific media partners, thus basically offloading fact checking to proven news publications. Others, like Reddit, weed out fake news via the community.
Tom Simonite wrote a piece on the many ways disabled people are being helped by machine learning. For example, Youtube recently rolled out algorithms that indicate applause, laughter, and music in captions. The company says user tests indicate that the feature significantly improves the experience of the deaf and hard of hearing (and anyone who needs to keep the volume down).
In addition, last year, Facebook launched a feature that uses the company’s research on image recognition to create text descriptions of images from a person’s friends, for example. Check out the article for more examples.
Smiths Detection Inc. (SDI) announced that it is partnering with the Duke University Edmund T. Pratt Jr. School of Engineering, Department of Electrical and Computer Engineering, in a “deep learning” digital solution project to advance airport checkpoint x-ray system screening capabilities.
Dan Gelston, President of SDI, said, “We must continue to invest in digital solutions to remain at the forefront of technology. This partnership, combined with our focus on innovation and experience in threat detection, leads the security industry in the development of state-of-the-art methods to help make the world a safer place.”