Robopocalypse Report #19: Mail Delivering Drones! Starfish Killing Robots!?! And Machine Learning to Pick Fashion Models?!?!?!

With a drone, a Brit reinvents the time lapse videos.

In Spain, they used drones to monitor tree plantations.

And it had to happen, someone went fishing with a drone...and he actually caught something.

Drone startup Measure has gotten permission from the FAA to fly a huge fleet of drones, 324 of them!

Poway, California outside of San Diego has passed a temporary ordnance banning drones from its airspace.

Finland has drones already delivering the mail.

Wilbur-Ellis is testing drones for agriculture use as well.

As reported earlier, a chimp took out a drone with a stick and now the scientists discuss the implications of the chimp's action.

Terrafugia, a flying car startup, has requested an FAA waiver for a UAV version of its flying car, the TF-X.  If it does not get it, they are threatening to move its work outside of the US.  I have found when companies, especially startups, get into threatening mode, its normally because they are not going to get what they want because they are not willing to do the work and are throwing a temper tantrum.

A Japanese insurance company is targeting drones.

CNBC talks about why its so hard for the US to make rules for drones.

The self driving car's smaller sibling, the self driving golf cart, seems to be a near complete success in Singapore.

Uber REALLY needs the self driving car, well, now.

Google's Self Driving cars are having issues...with human drivers.

Self driving cars are being called a good investment.

Huffington Post has caught up to my concerns about hacking self driving cars.

UC Berkeley has had a breakthrough in lasers (and here) that will reduce the size, weight and cost of LIDARs, the primary sensor of self driving cars. 

With the help of noninvasion spinal stimulation and an exoskeleton, a completely paralyzed man has been able to voluntarily control his legs for the first time since becoming paralyzed.

Ekso has developed an upper body exoskeleton that will be going on sale in 2016.

The exoskeleton market is expected to grow to $2.1 billion by 2021.

In general robotics, while not exactly with an Austrian accent or the Terminator Franchise,  a starfish killing robot is being tested at the Great Barrier Reef.

Also in the sea, scientists have demonstrated a type of floating robot called a MERMAID, for listening for undersea earthquakes.

The Australian brick laying bot, SAM, can do three times as much work as a human bricklayer and currently costs a half million dollars.

 UC Berkeley has developed a preschool (no, not kidding) for robots.

A Florida surgeon discusses robotic assisted surgery.

In the world of 3d printing, a new biosilk ink could make for innovative tissue printing.

MIT has come up with a way of simplifying the creation of CAD models for 3d printing.

HP is pushing hard into the 3d printing market.

ALCOA has decided to invest heavily in additive manufacturing (aka 3d printing).

 On the software bot side of the Robopocalypse, an AI program was written to create plot lines in interactive fiction (like video games).

An machine learning algorithm can now successfully predict whether or not someone's face will be successful as a fashion model.

In the economics side, VentureBeat looks at how drones will be transforming the economy.

ZDNet discusses how robotics is following IT and the internet's path to importance.

Oh the philosophy side, scientists and what-have-you are looking for information on major technological discontinuities from the past (with payments for accurate ones) where great change happened quickly.  They are looking for ways to avoid the shock of what true AI shows up.

On the humor side, a robot gives an interview and talks about taking over the world.  Take note, Noel.

BRETT: Berkeley Robot for the Elimination of Tedious Tasks

Researchers at the University of California, Berkeley, have developed algorithms that enable robots to learn motor tasks through trial and error using a process that more closely approximates the way humans learn, marking a major milestone in the field of artificial intelligence.

They demonstrated their technique, a type of reinforcement learning, by having a robot complete various tasks -- putting a clothes hanger on a rack, assembling a toy plane, screwing a cap on a water bottle, and more -- without pre-programmed details about its surroundings.

"What we're reporting on here is a new approach to empowering a robot to learn," said Professor Pieter Abbeel in UC Berkeley's Department of Electrical Engineering and Computer Sciences. "The key is that when a robot is faced with something new, we won't have to reprogram it. The exact same software, which encodes how the robot can learn, was used to allow the robot to learn all the different tasks we gave it."

Robopoclypse now.  ;)

link.

Pin Pointing new Potential Rodent Reservoirs for Disease Caused by Global Warming

Machine learning can pinpoint rodent species that harbor diseases and geographic hotspots vulnerable to new parasites and pathogens. So reports a new study in the Proceedings of the National Academy of Sciences led by Barbara A. Han, a disease ecologist at the Cary Institute of Ecosystem Studies.

Most emerging infectious diseases are transmitted from animals to humans, with more than a billion people suffering annually. Safeguarding public health requires effective surveillance tools.

Han comments: "Historically, emerging infectious diseases have been dealt with reactively, with efforts focused on containing outbreaks after they've spread. We were interested in how machine learning could inform early warning surveillance by revealing the distribution of rodent species that are effective disease reservoirs."

link.

Robopocalyptic Oceans! MIT Develops Machine Learning for Unmanned Underwater Vehicles

For the last decade, scientists have deployed increasingly capable underwater robots to map and monitor pockets of the ocean to track the health of fisheries, and survey marine habitats and species. In general, such robots are effective at carrying out low-level tasks, specifically assigned to them by human engineers -- a tedious and time-consuming process for the engineers.

When deploying autonomous underwater vehicles (AUVs), much of an engineer's time is spent writing scripts, or low-level commands, in order to direct a robot to carry out a mission plan. Now a new programming approach developed by MIT engineers gives robots more "cognitive" capabilities, enabling humans to specify high-level goals, while a robot performs high-level decision-making to figure out how to achieve these goals.

For example, an engineer may give a robot a list of goal locations to explore, along with any time constraints, as well as physical directions, such as staying a certain distance above the seafloor. Using the system devised by the MIT team, the robot can then plan out a mission, choosing which locations to explore, in what order, within a given timeframe. If an unforeseen event prevents the robot from completing a task, it can choose to drop that task, or reconfigure the hardware to recover from a failure, on the fly.

In March, the team tested the autonomous mission-planning system during a research cruise off the western coast of Australia. Over three weeks, the MIT engineers, along with groups from Woods Hole Oceanographic Institution, the Australian Center for Field Robotics, the University of Rhode Island, and elsewhere, tested several classes of AUVs, and their ability to work cooperatively to map the ocean environment.

The MIT researchers tested their system on an autonomous underwater glider, and demonstrated that the robot was able to operate safely among a number of other autonomous vehicles, while receiving higher-level commands. The glider, using the system, was able to adapt its mission plan to avoid getting in the way of other vehicles, while still achieving its most important scientific objectives. If another vehicle was taking longer than expected to explore a particular area, the glider, using the MIT system, would reshuffle its priorities, and choose to stay in its current location longer, in order to avoid potential collisions.

"We wanted to show that these vehicles could plan their own missions, and execute, adapt, and re-plan them alone, without human support," says Brian Williams, a professor of aeronautics and astronautics at MIT, and principal developer of the mission-planning system. "With this system, we were showing we could safely zigzag all the way around the reef, like an obstacle course."

link.

Why the Robopocalypse has Impaired Vision

Robopocalypse is Coming for Cooks: Baxter is Learning to Cook via Youtube

How’s this for a dystopian future: You finally receive your personal robot assistant, delivered to your door by Amazon drone. You unpack the shiny new machine, dust off the Styrofoam peanuts, and charge up the batteries. Then you switch it on and lead it to the kitchen so it can cook you dinner. The robot points its camera at you, waiting. Suddenly you realize in horror that your assistant doesn’t know how to cook, either—you’re supposed to teach it.

To prevent this nightmare dinnertime scenario, computer scientists are working on a robot that can teach itself to cook. It learns by watching YouTube videos.

This is much harder for a robot than it is for you, no matter how inept a cook you are. Imagine a mind that’s stumped by CAPTCHAs (“Letters with a squiggle through them? I’m out!”) trying to follow a video host who’s chatting and chopping at the same time. To tackle the task, University of Maryland graduate student Yezhou Yang and his coauthors broke it down into a few simpler pieces.

First, their robot would look at the person’s hands. For each hand, it would decide what type of grip the person was using. Was it a powerful grasp, as when holding a knife or a jar lid? Or was it a more delicate, precise grasp, maybe to lift a slice of bread from the counter? How wide was the object? The scientists taught the robot to recognize six grasps in all.

Next, the robot would try to identify the objects in the video. The researchers taught it 48 objects, including tools (such as spatula, bowl, and brush) and foods (meat, lettuce, yogurt, and so on).

After the robot had matched what the video host was holding in each hand came the crucial step—actually doing something.

“Due to the huge variation in human actions,” Yang says, it’s not yet possible for the robot to deduce what someone’s doing just by watching. So the researchers taught their robot to guess instead. Given the objects in its hands, the robot picked the most likely verb from 10 options: cut, pour, transfer, spread, grip, stir, sprinkle, chop, peel, or mix?

The authors chose 88 cooking videos from YouTube and used most of them to train their robot. The last dozen video clips—each showing just one cooking action—were the robot’s final exam.

The aspiring robot chef performed pretty well. After watching the test videos, it chose the right kind of grasp about 90% of the time. It correctly identified the objects about 80% of the time, and did equally well at guessing the action. Some of its mistakes happened when the videos included objects it hadn’t been trained on. When it saw a person using a knife to slice tofu, for example, the robot guessed that it was supposed to slice up a bowl.

link.

make sure he stays focused on the right videos.  Otherwise you might end up with some funky results.

Canadians Unleash Robopocalypse on our Genomes...Starting With Studying Autism

In the decade since the genome was sequenced in 2003, scientists and doctors have struggled to answer an all-consuming question: Which DNA mutations cause disease?

A new computational technique developed at the University of Toronto may now be able to tell us.

A Canadian research team led by professor Brendan Frey has developed the first method for 'ranking' genetic mutations based on how living cells 'read' DNA, revealing how likely any given alteration is to cause disease. They used their method to discover unexpected genetic determinants of autism, hereditary cancers and spinal muscular atrophy, a leading genetic cause of infant mortality.

Their findings appear in today's issue of the leading journal Science.

Think of the human genome as a mysterious text, made up of three billion letters. "Over the past decade, a huge amount of effort has been invested into searching for mutations in the genome that cause disease, without a rational approach to understanding why they cause disease," says Frey, also a senior fellow at the Canadian Institute for Advanced Research. "This is because scientists didn't have the means to understand the text of the genome and how mutations in it can change the meaning of that text." Biologist Eric Lander of the Massachusetts Institute of Technology captured this puzzle in his famous quote: "Genome. Bought the book. Hard to read."

What was Frey's approach? We know that certain sections of the text, called exons, describe the proteins that are the building blocks of all living cells. What wasn't appreciated until recently is that other sections, called introns, contain instructions for how to cut and paste exons together, determining which proteins will be produced. This 'splicing' process is a crucial step in the cell's process of converting DNA into proteins, and its disruption is known to contribute to many diseases.

Most research into the genetic roots of disease has focused on mutations within exons, but increasingly scientists are finding that diseases can't be explained by these mutations. Frey's team took a completely different approach, examining changes to text that provides instructions for splicing, most of which is in introns.

Frey's team used a new technology called 'deep learning' to teach a computer system to scan a piece of DNA, read the genetic instructions that specify how to splice together sections that code for proteins, and determine which proteins will be produced.

Unlike other machine learning methods, deep learning can make sense of incredibly complex relationships, such as those found in living systems in biology and medicine. "The success of our project relied crucially on using the latest deep learning methods to analyze the most advanced experimental biology data," says Frey, whose team included members from University of Toronto's Faculty of Applied Science & Engineering, Faculty of Medicine and the Terrence Donnelly Centre for Cellular and Biomolecular Research, as well as Microsoft Research and the Cold Spring Harbor Laboratory. "My collaborators and our graduate students and postdoctoral fellows are world-leading experts in these areas."

Once they had taught their system how to read the text of the genome, Frey's team used it to search for mutations that cause splicing to go wrong. They found that their method correctly predicted 94 percent of the genetic culprits behind well-studied diseases such as spinal muscular atrophy and colorectal cancer, but more importantly, made accurate predictions for mutations that had never been seen before.

They then launched a huge effort to tackle a condition with complex genetic underpinnings: autism spectrum disorder. "With autism there are only a few dozen genes definitely known to be involved and these account for a small proportion of individuals with this condition," says Frey.

In collaboration with Dr. Stephen Scherer, senior scientist and director of The Centre for Applied Genomics at SickKids and the University of Toronto McLaughlin Centre, Frey's team compared mutations discovered in the whole genome sequences of children with autism, but not in controls. Following the traditional approach of studying protein-coding regions, they found no differences. However, when they used their deep learning system to rank mutations according to how much they change splicing, surprising patterns appeared.

"When we ranked mutations using our method, striking patterns emerged, revealing 39 novel genes having a potential role in autism susceptibility," Frey says.

link.

The Implications of Machine Learning, Both Good and bad

He goes off rails at 18:00 minutes.  The problem is those improvements in computers are not coming.  There's a reason most HPC sites are investing in "post Moore's law" technical groups to keep HPC alive at least until quantum computing: Moore's law dies in the next five years.

Logical Limitations for Machine Ethics?

Logical Limitations to Machine Ethics with Consequences to Lethal Autonomous Weapons

Authors:

Engleert et al

Abstract:

Lethal Autonomous Weapons promise to revolutionize warfare -- and raise a multitude of ethical and legal questions. It has thus been suggested to program values and principles of conduct (such as the Geneva Conventions) into the machines' control, thereby rendering them both physically and morally superior to human combatants.

We employ mathematical logic and theoretical computer science to explore fundamental limitations to the moral behaviour of intelligent machines in a series of "Gedankenexperiments": Refining and sharpening variants of the Trolley Problem leads us to construct an (admittedly artificial but) fully deterministic situation where a robot is presented with two choices: one morally clearly preferable over the other -- yet, based on the undecidability of the Halting problem, it provably cannot decide algorithmically which one. Our considerations have surprising implications to the question of responsibility and liability for an autonomous system's actions and lead to specific technical recommendations.

Soooo....does that imply PEOPLE are not Turing Machines???  *evil grin*

A New Collaborative Machine Learning Algorithm

Machine learning, in which computers learn new skills by looking for patterns in training data, is the basis of most recent advances in artificial intelligence, from voice-recognition systems to self-parking cars. It's also the technique that autonomous robots typically use to build models of their environments.

That type of model-building gets complicated, however, in cases in which clusters of robots work as teams. The robots may have gathered information that, collectively, would produce a good model but which, individually, is almost useless. If constraints on power, communication, or computation mean that the robots can't pool their data at one location, how can they collectively build a model?

At the Uncertainty in Artificial Intelligence conference in July, researchers from MIT's Laboratory for Information and Decision Systems will answer that question. They present an algorithm in which distributed agents — such as robots exploring a building — collect data and analyze it independently. Pairs of agents, such as robots passing each other in the hall, then exchange analyses.

In experiments involving several different data sets, the researchers' distributed algorithm actually outperformed a standard algorithm that works on data aggregated at a single location.

"A single computer has a very difficult optimization problem to solve in order to learn a model from a single giant batch of data, and it can get stuck at bad solutions," says Trevor Campbell, a graduate student in aeronautics and astronautics at MIT, who wrote the new paper with his advisor, Jonathan How, the Richard Cockburn Maclaurin Professor of Aeronautics and Astronautics. "If smaller chunks of data are first processed by individual robots and then combined, the final model is less likely to get stuck at a bad solution."

Campbell says that the work was motivated by questions about robot collaboration. But it could also have implications for big data, since it would allow distributed servers to combine the results of their data analyses without aggregating the data at a central location.

"This procedure is completely robust to pretty much any network you can think of," Campbell says. "It's very much a flexible learning algorithm for decentralized networks."

link.

Here Comes Skynet: DARPA Tackles Machine Learning

Machine learning – the ability of computers to understand data, manage results, and infer insights from uncertain information – is the force behind many recent revolutions in computing. Email spam filters, smartphone personal assistants and self-driving vehicles are all based on research advances in machine learning. Unfortunately, even as the demand for these capabilities is accelerating, every new application requires a Herculean effort. Even a team of specially-trained machine learning experts makes only painfully slow progress due to the lack of tools to build these systems.

The Probabilistic Programming for Advanced Machine Learning (PPAML) program was launched to address this challenge. Probabilistic programming is a new programming paradigm for managing uncertain information. By incorporating it into machine learning, PPAML seeks to greatly increase the number of people who can successfully build machine learning applications and make machine learning experts radically more effective. Moreover, the program seeks to create more economical, robust and powerful applications that need less data to produce more accurate results – features inconceivable with today’s technology.

“We want to do for machine learning what the advent of high-level program languages 50 years ago did for the software development community as a whole,” said Kathleen Fisher, DARPA program manager.

“Our goal is that future machine learning projects won’t require people to know everything about both the domain of interest and machine learning to build useful machine learning applications. Through new probabilistic programming languages specifically tailored to probabilistic inference, we hope to decisively reduce the current barriers to machine learning and foster a boom in innovation, productivity and effectiveness.”

1.  DARPA is doing it.  That means its a DARPA hard problem and has a small percentage likelihood of suceeding.

 2.  I'm being flippant about skynet.  I don't think this will work. 

3.  I have always found it ... odd...interesting...weird...that folks believe that R&D for stuff like AI happens in places that are ones that stockpile weapons of any sort or develop anything of the sort either.