There are plenty of ways to manipulate photos to make you look better, remove red eye or lens flare, and so on. But so far the blink has proven a tenacious opponent of good snapshots. That may change with research from Facebook that replaces closed eyes with open ones in a remarkably convincing manner.
It’s far from the only example of intelligent “in-painting,” as the technique is called when a program fills in a space with what it thinks belongs there. Adobe in particular has made good use of it with its “context-aware fill,” allowing users to seamlessly replace undesired features, for example a protruding branch or a cloud, with a pretty good guess at what would be there if it weren’t. It does so with a Generative Adversarial Network, essentially a machine learning system that tries to fool itself into thinking its creations are real. In a GAN, one part of the system learns to recognise, say, faces, and another part of the system repeatedly creates images that, based on feedback from the recognition part, gradually grow in realism.
Bob Herman (Axios) writes about how It’s not unrealistic to think that 80% of what doctors do will be replaced by algorithms and artificial intelligence. The idea, evangelised by venture capitalist Vinod Khosla
two years ago, is that machines can more accurately diagnosis us — and that will reduce deadly medical errors and free doctors up to do other things.
How it works: Khosla sees a day where people will have an “AI physician to answer their questions — that will be free. Just like Google Maps is free.”
That could be an app on our phones. But this goes beyond Googling symptoms.
Imagine going to a doctor appointment or getting an MRI.
Your human doctor takes photos, enters notes or waits for the scan results, and then plugs everything into an AI platform.
AI startup Eigen Technologies has raised £13M from Goldman Sachs Principal Strategic Investments and Singapore’s state investment fund Temasek.
London-based Eigen uses artificial intelligence technology to read legal and financial documents, making it easier for lawyers and bankers to analyse complex contracts — everything from derivatives to land deeds — and find specific clauses. Goldman Sachs is a customer, as are Linklaters, Evercore, and ING.
A new survey by Pindrop of 500 IT and business decision-makers in the U.S., France, Germany and the U.K. found that 28% are using voice technology with customers today and 84% expect to be using it in the next year. A full 94% expect to be using voice AI with customers within two years. That is an overwhelming trend by enterprises planning to adopt voice AI technologies.
The study found that 94% of managers believe that voice technology is an important driver of customer satisfaction and 88% believe it can deliver a competitive advantage. Only 57% believe it will reduce the cost of customer transactions.
Tessian of London announced that it has raised $13M in venture capital as the UK-based company seeks to automate the problem of enterprise email security. Formerly known by the more literal name CheckRecipient, the company uses machine learning to discover and fix misaddressed emails, a seemingly mundane hiccup that in fact causes big security headaches for companies when sensitive information falls into the wrong hands.
Tessian’s machine intelligence technology scours corporate email networks to analyse common patterns and detect anomolies. If it detects something unusual in an email someone is trying to send, it will send a warning suggesting the user check the contents.
Yet another bastion of human skill and intelligence has fallen to the onslaught of the machines. A new kind of deep-learning machine has taught itself to solve a Rubik’s Cube without any human assistance. The milestone is significant because the new approach tackles an important problem in computer science—how to solve complex problems when help is minimal.
Stephen McAleer and colleagues from the University of California, Irvine have pioneered a new kind of deep-learning technique, called “autodidactic iteration,” that can teach itself to solve a Rubik’s Cube with no human assistance. The trick that McAleer and co have mastered is to find a way for the machine to create its own system of rewards.
Ian Tucker (Guardian) writes about what else is coming down the self-driving road?
Insurers at the wheel: An Oxford University
startup, Oxbotica, proposes to solve the problem of liability in a collision involving autonomous vehicles by allowing insurers access to the vast amounts of data the car generates, even allowing them to control a car in real time if it detects a dangerous situation.
Ending random jams: A recent paper published in Transportation Researchfound that autonomous cars could bring about the end of congestion with no obvious explanation.
Motion sickness: Boston startup ClearMotion is working on shock absorbers that will counter the feeling of movement – thereby, it hopes, reducing the need for sick bags.