At TEDx Chennai 2010 I had the good fortune to meet Dr. Tara Thiagarajan, who heads Madura Microfinance. Tara blogs occasionally on issues peripheral to the business of microfinance. At Physics of Poverty, with her analytical background in neuroscience and complexity theory, she dissects questions about the meaning of socioeconomic development, and strives to bring those questions to the core rather than the periphery of the microfinance industry.
Her latest post, titled Productivity Line, attempts to reconceptualize the poverty line. She suggests that instead of dividing the world into segments based on their incomes, and then agreeing upon a reasonable income as the threshold, i.e. the poverty line, why not segment the world based on economic productivity?
Her post provoked a few questions about the relationship between production and consumption both at the social and the individual level. I note these below.
Firstly, I understand and acknowledge that taking a developmental design approach is more useful and socially sustainable than an approach that hand-holds the poorest of the poor just up to the threshold of the cycle of consumption.
1. What is an ideal ratio of producers to consumers in a society so that the society is economically sustainable (let's define economic sustainability as the capacity to diversify, grow and self-renew).
2. What is an ideal ratio of consumption to production in an individual's life? One could think of this ratio as an index of Marxist alienation and therefore a proxy to measure work-satisfaction or happiness.
3. In general, how do these ratios vary as a function of population size and other demographics as well as economic complexity (let's define economic complexity as the complexity of division of labor and the diversity of goods and services produced and consumed)? For instance, in a subsistence farming society where the economic complexity is relatively low, one could imagine that the ratio of producers to consumers is one, but that is scarcely an indicator of progress. Similarly in an extremely large and diversified society the ratio of consumption to production at an individual level is extremely high, but that is not an indicator of well being necessarily.
4. If one could measure these ratios from the demographic data of target communities receiving MFI, then one could go about defining bounds on these ratios and subsequently designing developmental interventions to optimize them.
Further Reading:
David Roodman's Open Book Microfinance Blog
The Institute for New Economic Thinking
A recent article comparing metrics of economic complexity using input-output measures
The building blocks of economic complexity: A white paper that applies complex network metrics to quantify macroeconomic complexity.
Sunday, July 31, 2011
Thursday, May 26, 2011
Naturalistic is a fate worse than a fate worse than death
Two years ago, I had written about neologisms in science. I am finally able to give a bad example of a neologism: "naturalistic". The term I am referring to has very little to do with the ideas of naturalism in philosophy of science, or the arts, but a lot more to do with stimuli used in studies of neuroimaging. "Naturalistic" is vaguely defined as: a laboratory stimulus that is an approximation of the stimuli encountered in the natural world. So a movie would be a naturalistic stimulus. It really is a niggling issue, but I have seen a certain hesitation in the scientific community to call these stimuli "natural stimuli", and a preference towards using "naturalistic stimuli".
"Natural" itself can be defined as similar to or pertaining to nature. So, are we to understand that "naturalistic" is then "similar to or pertaining to something that is similar to or pertaining to nature"?
A fate verse zan deth, I say.
"Natural" itself can be defined as similar to or pertaining to nature. So, are we to understand that "naturalistic" is then "similar to or pertaining to something that is similar to or pertaining to nature"?
A fate verse zan deth, I say.
Tuesday, April 12, 2011
Computational neuroscience vs Neuroinformatics
Below are some hastily sketched thoughts, by no means complete, on the distinction between computational neuroscience and neuroinformatics.
Computational Neuroscience: The field posits computational candidates for mechanisms by which the brain carries out a certain function. When we say computational candidates, we loosely talk about algorithms. I think algorithms have two or more theoretical aspects. I'll try to articulate those below.
First, the goal of the function performed by the brain must be articulated by a cost function. For example, if the goal is reaching out for an object and grasping it, then the cost function could minimize muscular effort, minimize the #neurons needed to encode the task, or minimize the error rate of the task assuming that it is performed several times. Sometimes, the cost function need not describe a very specific task such as grasping, but could describe a general organizing principle of the brain - such as minimize energy consumption, minimize the use of connective tissue, etc.
Second, the process by which the cost function is optimized must be articulated, keeping in mind that such a process must be feasible in the wet brain. The wet brain provides structural and functional bounds on what a candidate algorithm can and cannot do.
With these basic ingredients, the flavours then vary because the choice of level of description can be vastly different. Someone can talk about how ion channel ratios on the cell membrane are optimal for grasping, whereas someone else can talk about why the number of cortical areas devoted to grasping the brain is optimal. Both these optimalities could be treated computationally by using selective pressures during evolution, or selective pressures during brain
development as explanatory variables. To complicate matters further, optimality in the brain can be posited at the level of evolution, brain development, learning (plasticity), and adaptation.
Neuroinformatics: The field is concerned with issues of data analysis and visualization of neuroimaging data for human interpretation purposes. The algorithms applied here (such as ICA/CCA/ridge regression etc.) need not conform to any constraints posited by the wet brain. The field does not aspire to explain how the brain performs a certain function - it just aids the process of evidence accumulation, which is of course important for theoretical and CNS because otherwise we wouldn't have phenomena to explain and our theories cannot be validated. In this sense neuroinformatics is a tool for experimental neuroscience.
Computational Neuroscience: The field posits computational candidates for mechanisms by which the brain carries out a certain function. When we say computational candidates, we loosely talk about algorithms. I think algorithms have two or more theoretical aspects. I'll try to articulate those below.
First, the goal of the function performed by the brain must be articulated by a cost function. For example, if the goal is reaching out for an object and grasping it, then the cost function could minimize muscular effort, minimize the #neurons needed to encode the task, or minimize the error rate of the task assuming that it is performed several times. Sometimes, the cost function need not describe a very specific task such as grasping, but could describe a general organizing principle of the brain - such as minimize energy consumption, minimize the use of connective tissue, etc.
Second, the process by which the cost function is optimized must be articulated, keeping in mind that such a process must be feasible in the wet brain. The wet brain provides structural and functional bounds on what a candidate algorithm can and cannot do.
With these basic ingredients, the flavours then vary because the choice of level of description can be vastly different. Someone can talk about how ion channel ratios on the cell membrane are optimal for grasping, whereas someone else can talk about why the number of cortical areas devoted to grasping the brain is optimal. Both these optimalities could be treated computationally by using selective pressures during evolution, or selective pressures during brain
development as explanatory variables. To complicate matters further, optimality in the brain can be posited at the level of evolution, brain development, learning (plasticity), and adaptation.
Neuroinformatics: The field is concerned with issues of data analysis and visualization of neuroimaging data for human interpretation purposes. The algorithms applied here (such as ICA/CCA/ridge regression etc.) need not conform to any constraints posited by the wet brain. The field does not aspire to explain how the brain performs a certain function - it just aids the process of evidence accumulation, which is of course important for theoretical and CNS because otherwise we wouldn't have phenomena to explain and our theories cannot be validated. In this sense neuroinformatics is a tool for experimental neuroscience.
Tuesday, March 15, 2011
Deb Roy on wordling junior's life
Stumbled upon a fascinating longitudinal data-intensive, visualization-intensive series of ongoing work at Deb Roy's group at the MIT Media Lab which prompted me to share some quick thoughts. Watch it here:
The talk is brilliantly structured: starting with an emotional moment, walking through some stunning data visualizations, transitioning into a pitch for bluefinlabs (his startup) with more chutzpah, and ending with the personal yet transcendental.
Scientifically, the work itself, IMHO is meant to be treated as a glimpse into the kinds of hypothesis that can be tested, rather than a definitive statement on language acquisition patterns in children. Further, I've only watched the TED talk and 18 minutes is not enough time to point out caveats, that too obvious ones. I imagine that among the various controls, they would have in fact segmented and separately treated utterances of 'water' by adult to adult vs. by adult to child, since implicit and explicit learning presumably have different mechanisms. But I'll refrain from speculating further without digging deeper here.
Technically, managing parallel feeds of audiovisual data does not seem straightforward in the least. Further, segmenting humans from cluttered fisheye scenes or target words from natural speech, and 100 TB of it, despite 50 years of AI research, is still a pretty big deal. Besides, it's the first graph on a TED talk with error bars! Respect!
Commercially, social media is one big ticket application, but couldn't the same infrastructure be applied to support decisions in interviews or boardrooms, or evaluations in high end schools or creches? What else?
But what captured my fascination the most was the power of big data to accelerate discovery at an unprecedented rate.
On large datasets and fishing expeditions
Scientists are just beginning to appreciate the power of trawling the world for extremely large datasets and subsequently testing various hypotheses on small subsets of the data. This approach (sometimes derogatorily called a fishing expedition) is in stark contrast to classical scientific method where an apriori hypothesis dictates an experiment design and an analysis procedure. The LHC experiment is one such expedition to fish for postulated elementary particles.
To illustrate the power of the fishing expedition approach, imagine if Deb had decided apriori that he wanted to study word utterance length over time. In a conventional longitudinal experiment he might have chosen a subset of words used in natural conversation, asked several child and caregiver pairs to come into a studio for 1h/day and recorded their speech. He might then have analyzed the data, observed this U shaped phenomenon, and reported it in a journal about language acquisition, where it would have promptly gathered dust. Even potentially interested colleagues might think several times before replicating the study with a different set of words, or correlating the word utterance length with spatial context, since acquiring funding and approvals for a longitudinal study would be prohibitive. Instead, with Deb's fishing expedition dataset which might become publicly available someday, any armchair scientist with computational resources can ask their own creative follow up questions of the data with minimal entry barrier!
However, fishing expeditions have their downsides. First and most obvious perhaps is the risk involved: what if there are no fish to be found? In other words, what if a generic experiment design is not powerful enough to eliminate alternative hypotheses until the ones being tested are left standing? Second, what if something is found but it is hard to tell with any confidence whether that something is in fact, a fish? Put differently, does the subset of data relevant to a particular hypothesis being tested (such as a correlation between two variables) have enough statistical power to falsify its corresponding null hypothesis? Last (and perhaps the hardest to spot), fishing expeditions may encourage scientists to operate in a complete vacuum of hypotheses (as opposed to designing for a multiplicity of hypotheses).
What are some other fishing expeditions and their successes and failures? How is the Human Genome Project different from the LHC experiment?
The talk is brilliantly structured: starting with an emotional moment, walking through some stunning data visualizations, transitioning into a pitch for bluefinlabs (his startup) with more chutzpah, and ending with the personal yet transcendental.
Scientifically, the work itself, IMHO is meant to be treated as a glimpse into the kinds of hypothesis that can be tested, rather than a definitive statement on language acquisition patterns in children. Further, I've only watched the TED talk and 18 minutes is not enough time to point out caveats, that too obvious ones. I imagine that among the various controls, they would have in fact segmented and separately treated utterances of 'water' by adult to adult vs. by adult to child, since implicit and explicit learning presumably have different mechanisms. But I'll refrain from speculating further without digging deeper here.
Technically, managing parallel feeds of audiovisual data does not seem straightforward in the least. Further, segmenting humans from cluttered fisheye scenes or target words from natural speech, and 100 TB of it, despite 50 years of AI research, is still a pretty big deal. Besides, it's the first graph on a TED talk with error bars! Respect!
Commercially, social media is one big ticket application, but couldn't the same infrastructure be applied to support decisions in interviews or boardrooms, or evaluations in high end schools or creches? What else?
But what captured my fascination the most was the power of big data to accelerate discovery at an unprecedented rate.
On large datasets and fishing expeditions
Scientists are just beginning to appreciate the power of trawling the world for extremely large datasets and subsequently testing various hypotheses on small subsets of the data. This approach (sometimes derogatorily called a fishing expedition) is in stark contrast to classical scientific method where an apriori hypothesis dictates an experiment design and an analysis procedure. The LHC experiment is one such expedition to fish for postulated elementary particles.
To illustrate the power of the fishing expedition approach, imagine if Deb had decided apriori that he wanted to study word utterance length over time. In a conventional longitudinal experiment he might have chosen a subset of words used in natural conversation, asked several child and caregiver pairs to come into a studio for 1h/day and recorded their speech. He might then have analyzed the data, observed this U shaped phenomenon, and reported it in a journal about language acquisition, where it would have promptly gathered dust. Even potentially interested colleagues might think several times before replicating the study with a different set of words, or correlating the word utterance length with spatial context, since acquiring funding and approvals for a longitudinal study would be prohibitive. Instead, with Deb's fishing expedition dataset which might become publicly available someday, any armchair scientist with computational resources can ask their own creative follow up questions of the data with minimal entry barrier!
However, fishing expeditions have their downsides. First and most obvious perhaps is the risk involved: what if there are no fish to be found? In other words, what if a generic experiment design is not powerful enough to eliminate alternative hypotheses until the ones being tested are left standing? Second, what if something is found but it is hard to tell with any confidence whether that something is in fact, a fish? Put differently, does the subset of data relevant to a particular hypothesis being tested (such as a correlation between two variables) have enough statistical power to falsify its corresponding null hypothesis? Last (and perhaps the hardest to spot), fishing expeditions may encourage scientists to operate in a complete vacuum of hypotheses (as opposed to designing for a multiplicity of hypotheses).
What are some other fishing expeditions and their successes and failures? How is the Human Genome Project different from the LHC experiment?
Wednesday, October 27, 2010
Quality poverty in Indian higher education
Came across an interesting essay by Ashok Jhunjhunwala, championing quality in Indian higher education.
Here is his paper:
http://rtbi-iitm.in/Ashok/education_link.html
He argues that quality is lacking in higher education primarily because of underpaid teaching staff, thus making the vocation unattractive for the best and brightest. He claims that teachers can only be better paid (salaries tripled) through increasing fees since government budget for higher education cannot be doubled overnight. Then, he does a lot of analysis for building the case of increasing fees while keeping access open, through financial instruments and strong regulation to prevent profiteering institutions.
An interesting aspect, that he touches upon but does not elaborate, is the fundamental conflict of interest between primary/secondary education and higher education! Improving the quality of primary eduation will increase the demand for higher education, because lower drop-out rate and better exposure will mean that more teenagers are college ready. This would triple the current levels of 3 million new college entrants and put a strain on supply or quality of higher education.
I agree with him about how underpaid university lecturers are, and how difficult it is for academics in the west to even consider a return to Indian institutions. But it seems to me that he misses out on three potential pieces to the puzzle, that can shift the burden from the end user (i.e. fee paying students) to some other intermediary parties.
First, enabling third parties to enter the training and educational content domain can decrease the teacher:student ratio (e.g. 1:20 --> 1:40), enabling more students to pay for fewer teachers' wages.
Second, allowing lecturers to supplement their income through consulting gigs made easier through stronger industry-university relations can decrease the strain on the institutional wage bill. It will also offer short-term rather than long-term return for indian industries, making more companies willing to participate. Lecturers can also be given sufficient freedoms to be intrapreneurs in this regard, by identifying and creating partnerships in their field of expertise.
Third, western universities that depend to a large extent on their supply of graduate students from India, might be incentivized to fund entry-level university education in India. In this context too, university lecturers can be paid to create formal programs of student/researcher exchange, creating an alternate income stream.
Here is his paper:
http://rtbi-iitm.in/Ashok/
He argues that quality is lacking in higher education primarily because of underpaid teaching staff, thus making the vocation unattractive for the best and brightest. He claims that teachers can only be better paid (salaries tripled) through increasing fees since government budget for higher education cannot be doubled overnight. Then, he does a lot of analysis for building the case of increasing fees while keeping access open, through financial instruments and strong regulation to prevent profiteering institutions.
An interesting aspect, that he touches upon but does not elaborate, is the fundamental conflict of interest between primary/secondary education and higher education! Improving the quality of primary eduation will increase the demand for higher education, because lower drop-out rate and better exposure will mean that more teenagers are college ready. This would triple the current levels of 3 million new college entrants and put a strain on supply or quality of higher education.
I agree with him about how underpaid university lecturers are, and how difficult it is for academics in the west to even consider a return to Indian institutions. But it seems to me that he misses out on three potential pieces to the puzzle, that can shift the burden from the end user (i.e. fee paying students) to some other intermediary parties.
First, enabling third parties to enter the training and educational content domain can decrease the teacher:student ratio (e.g. 1:20 --> 1:40), enabling more students to pay for fewer teachers' wages.
Second, allowing lecturers to supplement their income through consulting gigs made easier through stronger industry-university relations can decrease the strain on the institutional wage bill. It will also offer short-term rather than long-term return for indian industries, making more companies willing to participate. Lecturers can also be given sufficient freedoms to be intrapreneurs in this regard, by identifying and creating partnerships in their field of expertise.
Third, western universities that depend to a large extent on their supply of graduate students from India, might be incentivized to fund entry-level university education in India. In this context too, university lecturers can be paid to create formal programs of student/researcher exchange, creating an alternate income stream.
Monday, May 24, 2010
Chaturashrama as a metaphor for personal and professional management
Chaturashrama, an ancient Hindu subtext (part of the Manusmriti) describes the four ideal stages of a human's (man's) life. Indeed, Chaturashrama has received a lot of flak for being patriarchal, for being the source of much determinism in modern Indian society, and for being But this post is not about such existing debates.
The four stages (or ashramas) are Brahmacharya (the stage of learning and preparation for life), Grihastha (the stage of taking responsibility and acquiring material, and emotional wealth, thus performing one's core evolutionary duty), Vanaprastha (the state of learning to let go of wordly comforts and spreading wisdom) and Sanyasa (the stage of isolated contemplation). Implicit in the Chaturashrama is that the path to each stage is through the previous. For instance, one cannot learn to let go in the right way unless one has been in complete control. Thus, these four stages represent four key functions of a person's journey through life: learning, taking control, letting go, and reflection.
While the four stages are prescribed for different periods in life, the modern world is seldom so linear. I'll proceed to claim that each stage is constantly present in our lives, and further, that each stage individually can provide both positive and negative feedback on the other. I'll argue that mastering the balance between these constraints or tensions can result in unprecedented success (spiritual, emotional, and or financial).
To be successful in the modern world, any entity (be it an individual direct the course of his/her personal life or a large organization) must continually cycle through these four functions every waking moment. To elaborate, one must continually
The G-V Balance, or how to DO well. Lessons in executive excellence
Mastering the tension between taking control and letting go (Grihastha-Vanaprastha balance), is a trait that can be seen in the best leaders of the world.
The B-S Balance, or how to THINK well. Lessons in strategic excellence
Likewise, closing the loop between learning and reflection (Brahmacharya-Sanyasa balance) can been seen in some of the world's best thinkers. Here learning refers to casting a wide net for ideas and knowledge and having the humility and curiosity to learn from anyone, while reflection refers to the courage to think on one's own reject good ideas at times, the vision to separate bad ideas from good ones, and the capacity to synthesize new knowledge.
[Footnote: Couldn't resist, but excellent thinking is mostly about knowing your way around B-S!]
The D-T Balance, or how to THINK by DOING and DO by THINKING
While the G-V balance deals with the issue of how to be a good doer [in management world, an executive; in the scientific world, an experimentalist], the B-S balance deals with how to be a good thinker [in the management world, a strategist; in the scientific world, a theorist]. And doubtlessly, acquiring each balance is a lesson in acquiring the other. By extension, it is possible to imagine that one always pursue excellence in thinking through excellence in doing and vice versa, leading us to study a third balance: the thinking-doing balance.
Mastering each of these balances is a book in itself and may be the subject of future posts!
The four stages (or ashramas) are Brahmacharya (the stage of learning and preparation for life), Grihastha (the stage of taking responsibility and acquiring material, and emotional wealth, thus performing one's core evolutionary duty), Vanaprastha (the state of learning to let go of wordly comforts and spreading wisdom) and Sanyasa (the stage of isolated contemplation). Implicit in the Chaturashrama is that the path to each stage is through the previous. For instance, one cannot learn to let go in the right way unless one has been in complete control. Thus, these four stages represent four key functions of a person's journey through life: learning, taking control, letting go, and reflection.
While the four stages are prescribed for different periods in life, the modern world is seldom so linear. I'll proceed to claim that each stage is constantly present in our lives, and further, that each stage individually can provide both positive and negative feedback on the other. I'll argue that mastering the balance between these constraints or tensions can result in unprecedented success (spiritual, emotional, and or financial).
To be successful in the modern world, any entity (be it an individual direct the course of his/her personal life or a large organization) must continually cycle through these four functions every waking moment. To elaborate, one must continually
- learn (update oneself about the state of the world)
- take control (make decisions based on the knowledge gathered and take responsibility for the consequences)
- trust and let go (be able to delegate responsibilities to those who step up and trust them to do it, as well as move on from unexpected failures)
- reflect (create new knowledge through reevaluation and contextualization of own experience and state of the world).
The G-V Balance, or how to DO well. Lessons in executive excellence
Mastering the tension between taking control and letting go (Grihastha-Vanaprastha balance), is a trait that can be seen in the best leaders of the world.
The B-S Balance, or how to THINK well. Lessons in strategic excellence
Likewise, closing the loop between learning and reflection (Brahmacharya-Sanyasa balance) can been seen in some of the world's best thinkers. Here learning refers to casting a wide net for ideas and knowledge and having the humility and curiosity to learn from anyone, while reflection refers to the courage to think on one's own reject good ideas at times, the vision to separate bad ideas from good ones, and the capacity to synthesize new knowledge.
[Footnote: Couldn't resist, but excellent thinking is mostly about knowing your way around B-S!]
The D-T Balance, or how to THINK by DOING and DO by THINKING
While the G-V balance deals with the issue of how to be a good doer [in management world, an executive; in the scientific world, an experimentalist], the B-S balance deals with how to be a good thinker [in the management world, a strategist; in the scientific world, a theorist]. And doubtlessly, acquiring each balance is a lesson in acquiring the other. By extension, it is possible to imagine that one always pursue excellence in thinking through excellence in doing and vice versa, leading us to study a third balance: the thinking-doing balance.
Mastering each of these balances is a book in itself and may be the subject of future posts!
Subscribe to:
Posts (Atom)
