Google

Winc

YouTube

Spotify

Scientific Sense Podcast

Saturday, August 10, 2013

Suspended knowledge

A recent paper in the journal of Physical Review Letters, showing a new analysis of the Cosmic Microwave Background Radiation and hypothesizing how the constructs from the existing model – such as dark energy and Neutrinos – could be sufficient to explain them, is symptomatic of the common experimental problem, in which results almost always tend toward ex. ante expectations. This is now common in most scientific disciplines including Physics, Medicine, Economics and others.

Experiments have been useful for most of the scientific pursuit of humans. However, we are fast approaching a regime that has a critical level of knowledge that cannot be improved by experiments. Empiricism and experimentalism have contributed significantly to the acceleration of knowledge in the past. However, the alarming rise seen in confirmation bias in experimental results, across scientific disciplines also point to a level of knowledge saturation bounded by experiments. This is akin to an environmental change and if the human brain is not able to adapt to this new state of knowledge, it stands high risk of stagnation, or worse extinction.

One way to get around this stagnation is to increase the number of experiments designed specifically to disprove expectations. As the value of confirming experiments decline for knowledge creation, experiments that look for yet to be defined needle in yet to be defined haystack have to dominate. The problem, of course, is even in those cases, ex. ante expectations of the haystack and the properties of the possible needle will be corrupted by the status quo. It is possible that humans are unable to make a transition to the next wave of knowledge creation as the current education and industrial systems are simply unable to adapt to the new realities.

It will be sad if we experiment our way through finer and finer measurements of the non-existent, in a state of suspended and stagnant knowledge.

Saturday, August 3, 2013

Social automobile

A recent study from Virginia Tech describes how unstructured textual information from online discussion forums and other social media can be mined to unearth vehicle defects. This, in turn, helps quality management professionals to predict impending problems, make decisions and proactively manage safety and quality. This is an area that merits further exploration.

There are two important vectors of possible future innovation here. First, portfolio information from social media, albeit being highly variable and unpredictable at individual level, contains valuable insights to aid decision support. One could imagine similar techniques to predict disease outbreaks and terrorist activities. It has been known that health and law professionals have been taking advantage of social media to support decisions. With improved technologies in unstructured text mining, these concepts can be further developed. In the long run, one has to believe that such technologies will become widely available to the consumers as well.

Second, it also showcases the importance of product quality for manufacturers as any lack of it will be internalized not just by the immediate customer but the populace at large, very quickly. For example, for automobile manufacturers, it is increasingly important that any possible quality issue is identified and corrected before the user of the vehicle knows and blogs about it. In essence, they have to build smart cars, able to self-diagnose problems and communicate them to the manufacturer. Additionally, the manufacturer has to be able to aggregate information at the fleet level to identify any looming quality issue before it affects the user adversely. We may be fast approaching the establishment of social channels for machine-to-machine communications. If manufacturers do not get ahead of this developing wave, they will be consumed by those who do.

It took thousands of years for humans to invent technologies to connect them all together, allowing the emergence of social intelligence. It will take less time for machines to do the same.

Friday, August 2, 2013

K-NEST

Recent simulation of NEST on a K Computer in Japan demonstrated a neural network close to 1% of a human brain. Although this is encouraging and point to the possibility of simulating an entire brain using exa-scale computers in the future, it also raises many questions as to the utility of programmatic simulation of the human brain. It has been a fascination for many ordinary men and women, for ages. A powerful quantum computer, housed in such an efficient space, standing ready to learn and forget, tantalized those who attempted to replicate it in-silica – but is has all been in vein.

Simulation of the human brain by traditional means, using raw computing power, is unlikely to produce insights into its workings. The brain is an efficient quantum computer, not something that possesses very high computing power and memory as measured by traditional computing metrics. Its specs are mediocre at best but it does have a highly sophisticated operating system, that is able to take advantage of its limited capabilities. In the absence of such powerful software, able to adapt to modular and learning apps, some able to reconfigure on demand, it will be less interesting. Failure of the operating system or some critical apps, render the brain incompetent quickly, regardless of its supposed hardware depth.

Sheer scaling of computing power and memory, although a straightforward experiment, is unlikely to yield the mysteries of this remarkable organ.

Monday, July 29, 2013

Free Marketability

A recent study in the International Journal of the Economics of Business shows that the European Climate Exchange (ECX) has been an efficient way to curb carbon emissions. Free markets, yet again, provide a simple and holistic way to price divisible and tradable benefits and dis-benefits to society. Free trading participants of a system are the most likely to impute efficient prices to outcomes and drive optimal policies. Ignorant politicians and paranoid scientists are unlikely to devise the best policies for society – worse, they make the situation worse.

More generally, society is in the best position to determine the price of an outcome for itself. Attempts at prescriptively driving policies either through the analysis of incomplete data or the adherence to political agendas by a few, will always result in worse outcomes. The basic idea that there is an efficient price to pollution is missing in these emotional debates. Since pollution affects society and its participants, only a large cohort of these participants, expressing freely could determine the cost of it.

There is a need to focus on two areas – first, market mechanisms, that will efficiently price any anticipated positive or negative outcomes to society, including pollution have to be prevalent with easy access to all participants and second, entities that manipulate prices have to be tried with the possibility of the highest penalty – they are engaged in crimes against humanity.

Friday, July 26, 2013

Unified, but not grand

A recent article in the European Physical Journal explains how the behavior of  a complex dynamical system could be explained by the superposition of simpler underlying systems. The work hypothesizes that special relativity and quantum-mechanical dynamics are mathematically identical to two interlocked processes operating at different energy levels.

Grand unification theories have been directionless for decades, attempting the impossible – some spawning particles and others energy fields, to plug the holes in improbable theories. They have been inventing strings, membranes and other such constructs, that would make a dress-maker blush but with little advancement in fundamental knowledge. The notion that complexity can be explained by superposition of simpler processes have been with us for decades as well, but that sounded too simplistic for the pretending geniuses behind ivy walls.

Simpler explanations to complex phenomena will always dominate even though it may not be sufficiently robust to prop up the ego of Nobel seekers.

Saturday, July 20, 2013

Parasitic evolution

Recent research from Vanderbilt University demonstrating that microbes in the gut of animals can substantially influence selection and thus evolution, is interesting. It implies that the outcome of evolution is based on a constrained optimization taking into account the characteristics and needs of multiple organisms in a single matrix. The interesting question is what dominates the process and how it has shaped evolutionary history.

Single cell organisms, amazingly robust and extremely flexible, have been dominating the Earth ever since they arrived, possibly hitching a ride on an asteroid. They shaped the environment to suit them and fundamentally changed the course of the blue planet. They meticulously designed carriers for them to survive in harsh conditions and thrive invisibly. Nothing less could have been expected of a species that travelled through interstellar space, billions of years ago.

Their hosts, complex and customized carrying boxes for this remarkable species are left with sheer astonishment as they study these superior biological entities.

Sunday, July 14, 2013

Excess power

A recent study in the Journal of Neuroscience replicates the process of daydreaming in a computer simulation. It is a reminder that the most important asset of humanity – 7 billion brains – is significantly underutilized. An evolutionary quirk endowed humans with a massive organ that finds itself with little to do most of the time. It can control and operate the most sophisticated robot on Earth with less than 10% of the available power. The rest, is simply wasted – and out of sheer frustration, it ventures into daydreaming and night dreaming.

Humans have been attempting to use excess computing power in interesting ways. The Internet finally connected islands of processing power but the use of available excess computing power in the network is still in its infancy. Computing power, however, is a trivial resource compared to brain power – that remains largely disconnected and wasted. Productivity of modern societies is likely highly correlated with its ability to connect brains and use wasted brain power to solve complex and interesting problems.

A technology, akin to the Internet, able to connect brains and utilize the available excess capacity, is needed for humans to take the next step. This is a societal problem – something that the “singularity” peddling technocrats are ill-equipped to understand.

Tuesday, July 9, 2013

A measure of goodness

As the world, fully segmented into countries, religions and belief systems, fight to prove and disprove, it is interesting to think about a measure of goodness. Here, actions clearly matter more than words, for the latter is costless. A good person contributes to the wealth of humanity to such an extent that optimization of the system will require the protection and nourishment of that entity at almost any cost. Goodness, then, is about understanding the universe, its interconnectedness and its ability to solve the future and forget the past. It is about shedding the pretense, capturing the thoughts of beautiful minds and extending them. It is about shunning meaningless fights and challenging apparently meaningful present.

Goodness is indeed rare – it is not about wealth, intelligence, status or strength. It is all about beautiful thoughts that connect constrained ideas for the benefit of all. It is not about showcasing own self, but rather appreciating the random ones. It is not about winning at any cost, but rather assuring that victory has meaning. It is not process but content and it is not status but contribution. It is the ability to see the whole and not manufacture the details of the components. It is not running for the meaningless destinations but crawling to the beacons of enigma. It is, indeed, about beautiful thoughts – with fading but powerful examples, such as Einstein and Gandhi – and as a precious few look back in such bewilderment that they could find minds of such inexplicable depth and capabilities, they hang their heads in shame.

How does one measure goodness, then? With asymmetric pay-off to the individual, most rational and ordinary men and women will shun choices of implementing goodness at her own cost. She will be forced to optimize with harsh constraints on her own tactical life, with a programmatic and irreversible end. Could ordinary humans be good enough? Would they put yet another brick in the wall of humanity? Would they extend the knowledge handed down from generations or would they simply give up?

The value of society is the sum total of its goodness – a metric that is clear to those who want to optimize and less so to others.