The groundwork for machine learning was laid down in the middle of last century. But increasingly powerful computers – harnessed to algorithms refined over the past decade – are driving an explosion of applications in everything from medical physics to materials, as Marric Stephens discovers.
A flamingo lives 40 years and a human being lives 90 years; a mouse lives two years and an elephant lives 60. Why? What determines the lifespan of a species? After analyzing nine species of mammals and birds, researchers at the Spanish National Cancer Research Center (CNIO) found a very clear relationship between the lifespan of these species and the shortening rate of their telomeres, the structures that protect the chromosomes and the genes they contain. The relationship is expressed as a mathematical equation, a formula that can accurately predict the longevity of the species. The study was done in collaboration with the Madrid Zoo Aquarium and the University of Barcelona.
“The telomere shortening rate is a powerful predictor of species lifespan,” the authors write in the prestigious journal Proceedings of the National Academy of Sciences (PNAS).
The study compares the telomeres of mice, goats, dolphins, gulls, reindeer, vultures, flamingos, elephants and humans, and reveals that species whose telomeres shorten faster have shorter lives.
Algorithms meant to spot hate speech online are far more likely to label tweets “offensive” if they were posted by people who identify as African-American.
AI systems meant to spot abusive online content are far more likely to label tweets “offensive” if they were posted by people who identify as African-American.
The news: Researchers built two AI systems and tested them on a pair of data sets of more than 100,000 tweets that had been annotated by humans with labels like “offensive,” “none,” or “hate speech.” One of the algorithms incorrectly flagged 46% of inoffensive tweets by African-American authors as offensive. Tests on bigger data sets, including one composed of 5.4 million tweets, found that posts by African-American authors were 1.5 times more likely to be labeled as offensive. When the researchers then tested Google’s Perspective, an AI tool that the company lets anyone use to moderate online discussions, they found similar racial biases.
Continue reading “Google’s algorithm for detecting hate speech is racially biased” »
The art of matchmaking has traditionally been the province of grandmas and best friends, parents, and even—sometimes—complete strangers. Recently they’ve been replaced by swipes and algorithms in an effort to automate the search for love. But Kevin Teman wants to take things one step further.
The Denver-based founder of a startup called AIMM has built an app that matches prospective partners using just what they say to a British-accented AI. Users talk to the female-sounding software to complete a profile: pick out your dream home, declare whether you consider yourself a “cat person,” and describe how you would surprise a potential partner.
Continue reading “AI could be your wingman—er, wingbot—on your next first date” »
Machine learning, introduced 70 years ago, is based on evidence of the dynamics of learning in the brain. Using the speed of modern computers and large datasets, deep learning algorithms have recently produced results comparable to those of human experts in various applicable fields, but with different characteristics that are distant from current knowledge of learning in neuroscience.
Using advanced experiments on neuronal cultures and large scale simulations, a group of scientists at Bar-Ilan University in Israel has demonstrated a new type of ultrafast artificial intelligence algorithms—based on the very slow brain dynamics—which outperform learning rates achieved to date by state-of-the-art learning algorithms.
In an article published today in the journal Scientific Reports, the researchers rebuild the bridge between neuroscience and advanced artificial intelligence algorithms that has been left virtually useless for almost 70 years.
This video is the ninth in a multi-part series discussing computing and the second discussing non-classical computing. In this video, we’ll be discussing what quantum computing is, how it works and the impact it will have on the field of computing.
[0:28–6:14] Starting off we’ll discuss, what quantum computing is, more specifically — the basics of quantum mechanics and how quantum algorithms will run on quantum computers.
Continue reading “What Is Quantum Computing (Quantum Computers Explained)” »
We tend to think of AI as a monolithic entity, but it has actually developed along multiple branches. One of the main branches involves performing traditional calculations but feeding the results into another layer that takes input from multiple calculations and weighs them before performing its calculations and forwarding those on. Another branch involves mimicking the behavior of traditional neurons: many small units communicating in bursts of activity called spikes, and keeping track of the history of past activity.
Each of these, in turn, has different branches based on the structure of its layers and communications networks, types of calculations performed, and so on. Rather than being able to act in a manner we would recognize as intelligent, many of these are very good at specialized problems, like pattern recognition or playing poker. And processors that are meant to accelerate the performance of the software can typically only improve a subset of them.
Continue reading “One chip to rule them all: It natively runs all types of AI software” »
The research and development of neural networks is flourishing thanks to recent advancements in computational power, the discovery of new algorithms, and an increase in labelled data. Before the current explosion of activity in the space, the practical applications of neural networks were limited.
Much of the recent research has allowed for broad application, the heavy computational requirements for machine learning models still restrain it from truly entering the mainstream. Now, emerging algorithms are on the cusp of pushing neural networks into more conventional applications through exponentially increased efficiency.
Given that going viral on the Internet is often cyclical, it should come as no surprise that an app that made its debut in 2017 has once again surged in popularity. FaceApp applies various transformations to the image of any face, but the option that ages facial features has been especially popular. However, the fun has been accompanied by controversy; since biometric systems are replacing access passwords, is it wise to freely offer up our image and our personal data? The truth is that today the face is ceasing to be as non-transferable as it used to be, and in just a few years it could be more hackable than the password of a lifetime.
Our countenance is the most recognisable key to social relationships. We might have doubts when hearing a voice on the phone, but never when looking at the face of a familiar person. In the 1960s, a handful of pioneering researchers began training computers to recognise human faces, although it was not until the 1990s that this technology really began to take off. Facial recognition algorithms have improved to such an extent that since 1993 their error rate has been halved every two years. When it comes to recognising unfamiliar faces in laboratory experiments, today’s systems outperform human capabilities.
Continue reading “How to Hack a Face: From Facial Recognition to Facial Recreation” »
Facebook has announced a breakthrough in its plan to create a device that allows people to type just by thinking.
It has funded a study that developed machine-learning algorithms capable of turning brain activity into speech
It worked on epilepsy patients who had already had recording electrodes placed on their brains to asses the origins of their seizures, ahead of surgery.
