Five Important Things from the Last Week
There's so much relevant news from the past week, I can't just focus on any one thing... so here are five of the most significant things to hit my radar in past week:
In ascending order of importance.
5. On Marginal Revolution: What are some unknown but incredibly important inventors? Why can't we get rid of the penny? And what is the moral basis of capitalism?
4. Lawrence Berkeley lab and Oxford University researchers developed a particle accelerator that takes electron beams and powers them up to a billion electron volts (1 GeV) in only 3.3 centimeters using a technology called laser wakefield acceleration. If these particle accelerators become popular and start to edge out conventional accelerators, then we'll both learn a lot more about particle physics, and put ourselves at greater risk for creating a stable strangelet. Doing a risk/benefit calculation is difficult because of uncertainty in the probabilities involved.
3. If all goes well, we may start running our automobiles on ceramic ultracapacitors which take us 500 miles on only $9 worth of electricity using a battery that recharges in 5 minutes. Eric from Digital Crusader did a few basic calculations and found that transferring that amount of energy would require a rate of 1.2 megawatts, much greater than anything seen in current home electronics systems. The inventors of this technology claim that one day it will completely replace the internal combustion engine.
2. The mouse brain has been mapped down to individual cells. It only cost $41 million to do. A 3D atlas of the common lab mouse brain can be found here. If we had computers a couple orders of magnitude more powerful than today, we could start trying to simulate that mouse's brain in a virtual environment. The success of the mouse brain mapping project is also a testimony to the success of high-level philanthropy. Paul Allen, co-founder of Microsoft, contributed $50 million to the project, more than it even needed. This Merkle paper is relevant as background.
1. Chris Phoenix proposes the creation of cubic micron DNA structures. Specifically, Chris proposed "solid molecular constructions, using DNA as a backbone, plus other arbitrary molecules precisely positioned within the volume. " He estimates that it would be possible to design one for $10 million to $100 million once the entire process is automated. The idea came out of a thought experiment about what would be possible with today's technology and only a "moderate amount of engineering".
The idea would be to build bricks that can independently manufacture other bricks, to produce a rudimentary DNA nanofactory. Less ambitiously, you could design bricks that perform specialized tasks, like breaking down garbage efficiently, and then mass-produce the bricks to perform that function. The power of the approach is that, with current technology, you can precisely specify the DNA structure within a cubic micron volume, making it possible to eventually build any structure that can be designed. Because the density of DNA is about 1.3 nm^3 per base pair, it would take about 500 million base pairs worth of DNA to fill a cubic micron space. At current DNA synthesis prices ($0.10 if it's your machine) that works out to $50 million/block, but the cost is rapidly falling.
Chris expounds a bit more on the concept here. Meanwhile, CRN argues "yes, it's coming soon".
Researchers Model Salience and Attention
Working from first principles in Bayesian probability theory and Shannon's theory of communication, two Southern California researchers have developed a mathematical theory of surprise - and how the brain perceives novelty, importance, or noteworthiness. Pierre Baldi of UC Irvine and Laurent Itti of the University of Southern California developed the theory working with agents in a digital environment, and confirmed their findings with eye-tracking experiments using human subjects viewing dynamic stimuli in a variety of contexts.
The theory was so successful that Baldi and Itti were recently awarded a $600,000 NSF grant to test its validity further.
Itti hails from a computational neuroscience lab which seeks not only to model the human brain (specifically, how it delegates attention), but develop mathematically optimal algorithms with problem-solving applications in "automatic target detection in cluttered natural scenes, video compression, autonomous robotic nagivation on land or under water, or animation of virtual agents". The result are algorithms that give a better bang-for-your-bit on certain attention tasks than the human brain does. Applied to vision compression, the attention delegation model was able to cut filesize in half by preserving only the information in the video judged to be salient.
Bayesian probability theory may be used in a context-independent way to judge the extent to which an incoming piece of information forces a rational agent to change his or her beliefs. Applied to attention studies, Bayesian models showed superior performance to artificially constructed models of salience or computations of Shannon entropy.
The ambitious goal of the project is to develop a model that naturally breaks down a pure information stream (like the sequence of bits comprising a video file) into "feature channels" which isolate salient features in the information such as color or shape. This has obvious applications in machine vision and other areas.
The researchers state: "At the foundation of our model is a simple theory which describes a principled approach to computing surprise in data streams. While surprise is not a new concept it had lacked a formal definition, broad enough to capture the intuitive meaning of the term, yet quantitative and computable… Beyond vision, computable surprise could guide the development of data mining, as it can in principle be applied to any type of data, including visual, auditory or text."
In news articles the theory is portrayed as a model of human brain activity, but in actuality it goes beyond being a model. By working from first principles, the theory offers a recipe for attention-delegation that surpasses the capabilities of the human brain. Early applications might be the integration of "computable surprise" algorithms into a heads-up-display for soldiers on the field. Scanning a cluttered scene more rapidly than a human would be capable, such a system could alert the soldier to potential threats in advance of their noticing independently. Further applications would be in advanced Artificial Intelligence capable of formulating plans and accomplishing real-world goals.