Category Archives: Science

Machine envy

By Philip Ball – The tools of science are so specialised that we accept them as a kind of occult machinery for producing knowledge. We figure that they must know how it all works. Likewise, histories of science focus on ideas rather than methods — for the most part, readers just want to know what the discoveries were.

Even so, most historians these days recognise that the relationship between scientists and their instruments is an essential part of the story. It isn’t simply that the science is dependent on the devices; the devices actually determine what is known. You explore the things that you have the means to explore, planning your questions accordingly.

Today, however, they have become symbols of prestige as never before. I have several times been invited to admire the most state-of-the-art device in a laboratory purely for its own sake, as though I was being shown a Lamborghini.

One of the dysfunctional consequences of this sort of attitude is that the machine becomes its own justification, its own measure of worth. Results seem ‘important’ not because of what they tell us but because of how they were obtained. more>

How to Avoid a Tech Counterrevolution

By Leonid Bershidsky – It’s easy to laugh at a juice squeezer produced by a relatively small startup, whose real competence is in making fancy fruit-and-vegetable packets. It’s not really problem-solving tech; it’s a money-raising gimmick.

The problem is deeper though. Musk, a slicker marketer than Zuckerberg, talks about initially releasing a technology that would help people with brain damage — from strokes, for example.

Facebook is talking about “sharing thoughts,” hitting precisely on the most worrying aspects of the nascent technology: Who wants to share uncensored thoughts, especially with a company that collects information about its users without explaining to them exactly what is harvested? Who wants to give a machine built by a corporate entity access to one’s brain?

Tasks that desperately need automation and tech solutions are narrow. Thinking smaller and applying resources and energy to narrow, specific problems could be a good chance to build trust before it disappears entirely. more>

Raising good robots

We already have a way to teach morals to alien intelligences: it’s called parenting. Can we apply the same methods to robots?
By Regina Rini – Philosophers and computer scientists alike tend to focus on the difficulty of implementing subtle human morality in literal-minded machines. But there’s another problem, one that really ought to come first. It’s the question of whether we ought to try to impose our own morality on intelligent machines at all. In fact, I’d argue that doing so is likely to be counterproductive, and even unethical. The real problem of robot morality is not the robots, but us.

Can we handle sharing the world with a new type of moral creature?

We like to imagine that artificial intelligence (AI) will be similar to humans, because we are the only advanced intelligence we know. But we are probably wrong. If and when AI appears, it will probably be quite unlike us. It might not reason the way we do, and we could have difficulty understanding its choices.

Plato’s student Aristotle disagreed. He thought that each sort of thing in the world – squirrels, musical instruments, humans – has a distinct nature, and the best way for each thing to be is a reflection of its own particular nature.

‘Morality’ is a way of describing the best way for humans to be, and it grows out of our human nature. For Aristotle, unlike Plato, morality is something about us, not something outside us to which we must conform. Moral education, then, was about training children to develop abilities already in their nature. more>


Updates from GE

I Machine, You Human: How AI Is Helping GE Build A Powerhouse Of Knowledge
By Tomas Kellner – Every fall, GE Global Research holds a scientific gathering called the Whitney Symposium highlighting the latest scientific trends. Last year the two-day event explored industrial applications of artificial intelligence. We sat down with Mark Grabb and Achalesh Pandey, two GE scientists looking for ways to apply AI to jet engines, medical scanners and other machines.

“We are starting to see significant performance increases from the combination of deep learning and reinforcement learning, where you have a human in the loop correcting the system,” Grabb said. “Once you build a smooth user experience and get the system going, people don’t even know they are correcting the AI along the way.”

At GE, we are writing software like Predix, which is the cloud-based operating system for machines that allows us to connect them to the Industrial Internet. But we also have a tremendous number of domain experts. There’s a lot of physics and domain knowledge that’s required to build good analytics and machine learning models. We have actually built AI systems that help data scientists more quickly and more effectively capture the domain knowledge across all the people inside GE building these models. So AI comes in even in the developing of analytics. more>

Updates from GE

Octopus And Squid Evolution Is Officially Weirder Than We Could Have Ever Imagined
By Signe Dean – They edit their own genes!

Just when we thought octopuses couldn’t be any weirder, it turns out that they and their cephalopod brethren evolve differently from nearly every other organism on the planet.

In a surprising twist, scientists have discovered that octopuses, along with some squid and cuttlefish species, routinely edit their RNA (ribonucleic acid) sequences to adapt to their environment.

This is weird because that’s really not how adaptations usually happen in multicellular animals. When an organism changes in some fundamental way, it typically starts with a genetic mutation – a change to the DNA.

Those genetic changes are then translated into action by DNA’s molecular sidekick, RNA. You can think of DNA instructions as a recipe, while RNA is the chef that orchestrates the cooking in the kitchen of each cell, producing necessary proteins that keep the whole organism going.

But RNA doesn’t just blindly execute instructions – occasionally it improvises with some of the ingredients, changing which proteins are produced in the cell in a rare process called RNA editing.

When such an edit happens, it can change how the proteins work, allowing the organism to fine-tune its genetic information without actually undergoing any genetic mutations. But most organisms don’t really bother with this method, as it’s messy and causes problems more often that solving them.

“The consensus among folks who study such things is Mother Nature gave RNA editing a try, found it wanting, and largely abandoned it,” Anna Vlasits reports for Wired.

But now it looks like cephalopods didn’t get the memo. more>

Now it’s time to prepare for the Machinocene


Expressivism, Pragmatism and Representationalism, Author: Huw Price.

By Huw Price – One way or another, then, we are going to be sharing the planet with a lot of non-biological intelligence. Whatever it brings, we humans face this future together. We have an obvious common interest in getting it right. And we need to nail it the first time round. Barring some calamity that ends our technological civilization without entirely finishing us off, we’re not going to be coming this way again.

If we are to develop machines that think, ensuring that they are safe and beneficial is one of the great intellectual and practical challenges of this century. And we must face it together – the issue is far too large and crucial to be tackled by any individual institution, corporation or nation. Our grandchildren, or their grandchildren, are likely to be living in a different era, perhaps more Machinocene than Anthropocene.

Our task is to make the best of this epochal transition, for them and the generations to follow. We need the best of human intelligence to make the best of artificial intelligence. more>

The Spiritual, Reductionist Consciousness of Christof Koch

By Steve Paulson – Consciousness is a buzzing business in neuroscience labs and brain institutes. But it wasn’t always this way. Just a few decades ago, consciousness barely registered as a credible subject for science.

Why were humans able to create civilizations that have transformed the planet?

We don’t have a precise answer. We have big brains and are, by some measure, the most intelligent species, at least in the short term. We’ll see whether we’ll actually survive in the long term, given our propensity for mass violence. And we’ve manipulated the planet to such an extent that we are now talking about entering a new geological age, the Anthropocene.

But it’s unclear why whales or dolphins—some of which have bigger brains and more neurons in their cortex than we do—why they are not called smarter or more successful. Maybe because they have flippers and live in the ocean, which is a relatively static environment. With flippers, you’re unable to build sophisticated tools.

Of course, human civilization is all about tools, whether it’s a little stone, an arrow, a bomb, or a computer. more>


Updates from GE

Game On: Augmented Reality Is Helping Factory Workers Become More Productive
By Tomas Kellner – As chief engineer for advanced manufacturing at GE Healthcare, Jimmie Beacham, 43, is in charge of a futuristic laboratory in Waukesha, Wisconsin, experimenting with new ways to make things. He and his team are using the Xbox and a connected Kinect motion tracker to bring augmented reality (AR) into the factory and help workers become more efficient. “We are projecting the work instructions onto the parts and use sensors to monitor the assembly and give feedback to the operator,” Beacham says.

Specifically, the Kinect and a camera are following the worker’s movements and feeding them to a computer that stores the assembly instructions. The computer controls an overhead projector that displays the manufacturing steps on the workbench. Based on the visual and sensory feedback, the system signals the operator immediately if an error occurs or guides them to the next step.

The system currently at the Waukesha lab came from Light Guide System, a Detroit-area maker of augmented reality tools for industry. The first applications are focusing on guiding workers through “the critical steps where we can’t afford to make a mistake,” Beacham says. But his team has already started expanding its scope and connecting it to face recognition technology, collaborative robots, or cobots, and Predix, GE’s software platform for the Industrial Internet. more>

Turning Back Time: Watching Rust Transform into Iron

By W. Zhu, J.P. Winterstein, W.D. Yang, L. Yuan, R. Sharma and G. Zhou – Using a state-of-the-art microscopy technique, experimenters at the National Institute of Standards and Technology (NIST) and their colleagues have witnessed a slow-motion, atomic-scale transformation of rust—iron oxide—back to pure iron metal, in all of its chemical steps.

In a new effort to study the microscopic details of metal oxide reduction, researchers used a specially adapted transmission electron microscope (TEM) at NIST’s NanoLab facility to document the step-by-step transformation of nanocrystals of the iron oxide hematite (Fe2O3) to the iron oxide magnetite (Fe3O4), and finally to iron metal.

By lowering the temperature of the reaction and decreasing the pressure of the hydrogen gas that acted as the reducing agent, the scientists slowed down the reduction process so that it could be captured with an environmental TEM—a specially configured TEM that can study both solids and gas. The instrument enables researchers to perform atomic-resolution imaging of a sample under real-life conditions—in this case the gaseous environment necessary for iron oxides to undergo reduction–rather than under the vacuum needed in ordinary TEMs. more>

3 Ways Exponential Technologies are Impacting the Future of Learning

By Sveta McShane – Exponential technologies have a tendency to move from a deceptively slow pace of development to a disruptively fast pace. We often disregard or don’t notice technologies in the deceptive growth phase, until they begin changing the way we live and do business. Driven by information technologies, products and services become digitized, dematerialized, demonetized and/or democratized and enter a phase of exponential growth.

In the past three decades, jobs requiring routine manual or routine cognitive skills have declined as a percent of the labor market. On the other hand, jobs requiring solving unstructured problems, communication, and non-routine manual work have grown.

The best chance of preparing young people for decent paying jobs in the decades ahead is helping them develop the skills to solve these kinds of complex tasks. more>