Computational Journalism — Google awards to projects around computational journalism. Sample: The goal of the project is to automatically build topic “event threads” that will help journalists and citizens decode claims made by public figures, in order to distinguish between personal opinion, communication tools, and voluntary distortions of the reality.
Editing the Software of Life — research yielding the ability to edit DNA has spawned a new set of biotech startups, and a patent morass. Zhang already holds the first of several vital and broad patents covering cas9 genome editing. Yet, Doudna and Charpentier had filed patent applications covering similar ground earlier than Zhang.
Molecular Programming Project — aims to develop computer science principles for programming information-bearing molecules like DNA and RNA to create artificial biomolecular programs of similar complexity. Our long-term vision is to establish molecular programming as a subdiscipline of computer science — one that will enable a yet-to-be imagined array of applications from chemical circuitry for interacting with biological molecules to nanoscale computing and molecular robotics.
The Software Analysis Workbench — provides the ability to formally verify properties of code written in C, Java, and Cryptol. It leverages automated SAT and SMT solvers to make this process as automated as possible, and provides a scripting language, called SAW Script, to enable verification to scale up to more complex systems. “Non-commercial” license.
What’s Wrong with Deep Learning? (PDF in Google Drive) — What’s missing from deep learning? 1. Theory; 2. Reasoning, structured prediction; 3. Memory, short-term/working/episodic memory; 4. Unsupervised learning that actually works. … and then ways to get those things. Caution: math ahead.
Psychology of Software Architecture — a wonderful piece of writing, but this stood out: It comes down to behavioral economics and game theory. The license we choose modifies the economics of those who use our work.
Internet Users Increasingly Blocking Ads, Including on Mobiles (The Economist) — mobile networks working on ad blockers for their customers, If lots of mobile subscribers did switch it on, it would give European carriers what they have long sought: some way of charging giant American online firms for the strain those firms put on their mobile networks. Google and Facebook, say, might have to pay the likes of Deutsche Telekom and Telefónica to get on to their whitelists.
Connasence (Wikipedia) — a taxonomy of (systems) coupling. Two components are connascent if a change in one would require the other to be modified in order to maintain the overall correctness of the system. (Via Ben Gracewood.)
Unpowered Ankle Exoskeleton — “As we understand human biomechanics better, we’ve begun to see wearable robotic devices that can restore or enhance human motor performance,” says Collins. “This bodes well for a future with devices that are lightweight, energy-efficient, and relatively inexpensive, yet enhance human mobility.”
Body-Powered Processing (Ars Technica) — The new SAM L21 32-bit ARM family of microcontroller (MCUs) consume less than 35 microamps of power per megahertz of processing speed while active, and less than 200 nanoamps of power overall when in deep sleep mode—with varying states in between. The chip is so low power that it can be powered off energy capture from the body. (via Greg Linden)
The Shut-In Economy — In 1998, Carnegie Mellon researchers warned that the Internet could make us into hermits. They released a study monitoring the social behavior of 169 people making their first forays online. The Web-surfers started talking less with family and friends, and grew more isolated and depressed. “We were surprised to find that what is a social technology has such anti-social consequences,” said one of the researchers at the time. “And these are the same people who, when asked, describe the Internet as a positive thing.”
Remotely Bricking Cars (BoingBoing) — story from 2010 where an intruder illegally accessed Texas Auto Center’s Web-based remote vehicle immobilization system and one by one began turning off their customers’ cars throughout the city.
Machine Learning Classification over Encrypted Data (PDF) — It is worth mentioning that our work on privacy-preserving classification is complementary to work on differential privacy in the machine learning community. Our work aims to hide each user’s input data to the classification phase, whereas differential privacy seeks to construct classifiers/models from sensitive user training data that leak a bounded amount of information about each individual in the training data set. See also The Morning Paper’s unpacking of it.
Privacy of Phone Audio (Reddit) — unconfirmed report from Redditor I started a new job today with Walk N’Talk Technologies. I get to listen to sound bites and rate how the text matches up with what is said in an audio clip and give feed back on what should be improved. At first, I though these sound bites were completely random. Then I began to notice a pattern. Soon, I realized that I was hearing peoples commands given to their mobile devices. Guys, I’m telling you, if you’ve said it to your phone, it’s been recorded…and there’s a damn good chance a 3rd party is going to hear it.
stenographer (Google) — open source packet dumper for capturing data during intrusions.
Which GPU for Deep Learning? — a lot of numbers. Overall, I think memory size is overrated. You can nicely gain some speedups if you have very large memory, but these speedups are rather small. I would say that GPU clusters are nice to have, but that they cause more overhead than the accelerate progress; a single 12GB GPU will last you for 3-6 years; a 6GB GPU is plenty for now; a 4GB GPU is good but might be limiting on some problems; and a 3GB GPU will be fine for most research that looks into new architectures.
23andMe Wins FDA Approval for First Genetic Test — as they re-enter the market after FDA power play around approval (yes, I know: one company’s power play is another company’s flouting of safeguards designed to protect a vulnerable public).
Real World Active Learning — the point at which algorithms fail is precisely where there’s an opportunity to insert human judgment to actively improve the algorithm’s performance. An O’Reilly report with CrowdFlower.
Hearing With Your Tongue (BoingBoing) — The tongue contains thousands of nerves, and the region of the brain that interprets touch sensations from the tongue is capable of decoding complicated information. “What we are trying to do is another form of sensory substitution,” Williams said.
Program Synthesis Explained — The promise of program synthesis is that programmers can stop telling computers how to do things, and focus instead on telling them what they want to do. Inductive program synthesis tackles this problem with fairly vague specifications and, although many of the algorithms seem intractable, in practice they work remarkably well.
Ev Williams on Metrics — a master-class in how to think about and measure what matters. If what you care about — or are trying to report on — is impact on the world, it all gets very slippery. You’re not measuring a rectangle, you’re measuring a multi-dimensional space. You have to accept that things are very imperfectly measured and just try to learn as much as you can from multiple metrics and anecdotes.
Nature, the IT Wizard (Nautilus) — a fun walk through the connections between information theory, computation, and biology.
Smartest Cities Rely on Citizen Cunning and Unglamorous Technology (The Guardian) — vendors like Microsoft, IBM, Siemens, Cisco and Hitachi construct the resident of the smart city as someone without agency; merely a passive consumer of municipal services – at best, perhaps, a generator of data that can later be aggregated, mined for relevant inference, and acted upon. Should he or she attempt to practise democracy in any form that spills on to the public way, the smart city has no way of accounting for this activity other than interpreting it as an untoward disruption to the orderly flow of circulation.
Introduction to the Modern Brain-Computer Interface Design (UCSD) — The lectures were first given by Christian Kothe (SCCN/UCSD) in 2012 at University of Osnabrueck within the Cognitive Science curriculum and have now been recorded in the form of an open online course. The course includes basics of EEG, BCI, signal processing, machine learning, and also contains tutorials on using BCILAB and the lab streaming layer software.