Safety First: Stochastic Decision Problems

Vladimir Norkin of the Glushkov Institute of Cybertronics in Kiev gave a very interesting (and very equation based) seminar on mixed integer approaches in stochastic optimisation on Friday afternoon. Norkin has been a visitor in IME for the last week. Titled "Safety First Class of Stochastic Decision Problems: Mixed Integer Approach" he explored examples of portfolios with critical thresholds.

Portfolio selection is a relatively new area for me, something I have learnt since arriving at IIASA. Its interesting to see how the idea is used in a variety of sectors and the philosophy of risk minimisation by selecting a diverse portfolio. Safety first means the minimisation of the probability of bad events.

Norkin started the seminar with a quote from Ulrich Beck (1986) that "society starts to realise its exposure to changes including those that appear due to its own development..." and then moved on to AD Roy's (1952) paper on "Safety first and the holding of assets". Its interesting to note that while mathematical techniques and computational capabilities have increased since then, the idea and fundamental theory still holds true.

Norkin also discussed techniques associated with a finite scenario set that relates directly to my YSSP project. He intimated that all finite scenario set probability optimisation problems can be reduced to a mixed Boolean programming problem building on Korbut and Finkelstein (1969).

PPA: Implementation and Evaluation

This seminar really should be title "Mind the Gap" which fits perfectly with the new overall course title "How to analyse the Ministry of Magic". Public Policy Analysis seminars offer a vast range of odd examples

Today we discussed the Implementation and Evaluation of policies and tools available to facilitate this. There are two approaches; Top-down and Bottom-up that cover the horizontal and vertical policy space.

The top-down approach is the one I am most familiar approach. A policy is formulated by a high-power (government? Ministry of Magic?) and then the orders for implementation are filtered down a change of command. This can lead to a Broken Telephone effect with information leaking at each step (Is it geeky to admit that I like the visual of heat equations that lose energy in each direction?). How can one improve this? Perhaps offer implementation incentives, improve communication channels. This is an ideal type, there is no policy that fits perfectly into this approach but it is interesting to note why policies failed using it such as poor formulation or power issues. The analysis can show the implementation gap...or Mind the Gap!

The bottom-up approach seems to be a bit more haphazard in nature which conflicts with my personality. The policy is dependent on a network of street-level bureaucrats that share resources and have a level of discretion. How do you analyse this? With great difficulty, there is a horizontal layer with weak structure from what I can gather. Policies are negotiated and there is a great deal of uncertainty. Policies have very vague goals with no associated metrics.

What have I learnt from this 6 seminar course dip into the world of social science? I am very much a quantitative scientist who likes structure, people are confusing and politics even more so. Do I ever want to enter the minefield of policy design, probably not. I think I'd offend too many people by telling them their process sucks.

Suggested Reading:
Paul Sabatier (1986), Top-down and Bottom-up Approaches to Implementation Research: A Critical Analysis and Suggested Synthesis”,Journal of Public Policy, Vol. 6, No. 1. (Jan. - Mar., 1986), pp. 21-48.

PPA: Decisions: Power and Rationality

Friday our ever decreasing group of public policy students debated the process of making decisions and what types of rationality exist. As the penultimate seminar we have almost completed the policy analysis cycle. Our reading list for the seminar included Lindblom and Dahl. Dahl argues that a key characteristic of a democracy is the continuing responsiveness of the government to the preferences of its citizens, considered as political equals. A political system which has the characteristic of being almost or completely responsive to all its citizens.

You can for the sake of argument and analysis divide most governments in four categories based on Contestation (C) and Inclusion (I). High C / High I, High C/ Low I, Low C/ High I, Low C/Low I. From this can further analyse with respect to:

• How the power is distributed
• Role of the state
• Nature of conflict
• Whose interests prevails
• Basis of power

The other topic up for debate was Rationality. Now this interests me, I've been reading up on rationality in decision-making for modeling purposes. The two we explored were Incrementalism (Partisan Mutual Adjustment?) and Bounded Rationality.

I'm really interested in Bounded Rationality and how people cannot considered more than a certain amount of information, important when asking decision-makers to pick based on multiple criteria or systems with multiple implications.

If by this point you're wondering when I became a social science convert I can quite adamantly profess to this being my only foray into the world of qualitative, mumble-jumble in which there are few wrong answers...I'll even take statistics equations over this world.

Reading List:

• Charles Lindblom (1950), “The Science of Muddling Through”, Public Administration Review, (19), Spring 1950, pp.79-88.
• Robert Dahl (1971), Polyarchy: Participation and Opposition, Yale University Press, Yale., Chapter 1: “Democratisation and Public Opposition”

PPA: Social Problems

We attended the third Public Policy Lecture today on Social problems. This is the first week I have actually enjoyed the readings, they made logical sense (I know its a very scientific view). Policy analysis thinks about solving problems (policies are MEANT to solve problems). There are many problems and beliefs as to why they are in fact social problems. Good examples include drug abuse, guns, poverty, energy efficiency. Each has a state in the world and is measurable by some metric.

In framing a problem there are three (well more but the are only looking at three) viewpoints. Positivism, Functionalism and Social Constructivism. When reviewing the problem one looks at the Nature of the problem, Causes of social problem, Methods to review and ways to Intervene.

In Positivism (exogenous), the problem is considered a disease caused by individuals or societies failings (weaknesses). One can measure through observation and remove the cause (excise disease). This approach is expert-driven and undemocratic.

The second Functionalism (endogenous) looks at a problem as a function to society. One needs deviants to know normality. One can once again measure this and then adjust norms to maintain a balance. This approach is also expert-driven.

The last is Social Constructivism and in this you can re-interpret the problem and states that there is no direct access to reality, that everything we do is an interpretation. It believes that social constructions reflect conflicts. Unlike the previous two Social Constructivism is not analytic but rather hermeneutic (interpretive). It challenges dormant constructions. This is one viewpoint I have the most difficulty understanding.

This is a reference that came up during the lecture: Charles Booth who designed the first poverty maps of London, his scale cause a lot of comic relief.

The associated readings:

• Fuller and Myers (1941), “A Natural History of Social Problems”, American Sociological Review, 6:3, pp.320-329.
• Elaine B. Sharp (1994), “Paradoxes of National Antidrug Policymaking”, in David A. Rochefort and Roger W. Cobb (1994), The Politics of Problem Definition: Shaping the Policy Agenda, University Press of Kansas, Kansas. pp. 98-116

IME: Coping with Uncertainty: Spatio-temporal heterogeneities

The second uncertainty seminar dealt with spatio-temporal problems. This seminar series is not meant to teach us new techniques but to create an awareness that one shouldn't automatically use the stock standard stochastic tools available like averaging, regression etc.

When analysing spatio-temporal problems they often have multiple scales, high dimensionality and multimodal dimensions. There are methods available to deal with this, most notably discounting and Fast Monte Carlo simulations. Consider the scatter plot above, its multimodal. How do you describe the multi-dimensional density function?
The Adaptive Monte Carlo STO procedure can be used for optimization module (structural and non-structural decision, production allocation).

All these techniques are all of interest to me currently. I'm currently designing a decision framework and to cope with the uncertainty one needs to implement some sort of robust solution. Note the some sort, I haven't got there yet.

Hydrogen Economy

I skipped today's lecture by Cesare Marchetti in favour of work but apparently I missed an never to be forgotten blend of science and fiction. The topic: "How the Hydrogen Economy did pop up in the 60s and how it will fill the world in 2060" was just the start.

I've pasted part of the abstract as a case in point: "By producing hydrogen from water, oxygen would be released and travel by air to the point where hydrogen is consumed. You don't need the second wire to bring back the electrons...From these explorations it came out that hydrogen could efficiently cover the whole of the second half of the energy budget and consequently, at least conceptually, nuclear energy could become the sole primary energy for humanity...So I introduced thermochemical processes where a set of chemical cyclic reactions driven by nuclear heat led finally to the decomposition of water into hydrogen and oxygen...where the big strategy was deployed, converging finally in the concept of the Energy Island where technology is imported in an atoll where giant rafts with nuclear reactors and chemical plants are moored. Uranium is extracted from the seawater used for cooling and fission products put in capsules that selfsink by melting the rocks, down to 20 km."

I wait with bated breathe for this solution to work...

P.S. If you want a review of the actual lecture visit Elisabeth's only English blog post of the YSSP.

IME: Modeling Primer

Yesterday afternoon saw the second seminar of the day, 'Modeling Primer' by the IME group. This lecture delved into the world of rational decision making: making the right decision at the right time. The right decision takes experience, intuition and knowledge (facts and rules).

It is a myth that it is easy to find a solution to a complex problem, for simple cases the decision is correct but you are taking a chance for complex problems.

A model is an adequate (for a purpose) representation of a part of (interdisciplinary) knowledge. I reiterate a quote from the beginning of June: All models are wrong but some are useful. They provide a tool to HELP find the best decision. BEST is merely a suggestion of decisions for reaching certain goals which can be done through analysis of trade-offs between conflicting goals or an evaluation of the consequences of various decisions.

A model requires consistency, transparency, replicability and documentation. Something to remember is that RARE events do happen, the low probability mean it won't happen.

Some parting words of wisdom from A. Einstein - "Everything should be made as simple as possible, but not one bit simpler." Words to live by.

Science on Global Problem Area: Poverty & Equity

Today's seminar on Science on Global Problem Areas focuses on Poverty and Equity. This topic relates specifically to southern Africa and there are many questions and few answers. There are a few strategic research questions and themes presented by HGC, POP and ENE:

Understanding Poverty:

• Systems analytical approaches to poverty (different dimensions of poverty and their linkages)
  
• Implications of population and social dynamics on risk and vulnerabilities
• Welfare benefits of improved access to basic good, services and technologies

Understanding Equity:

• Analysing distributions;schemes for allocating scarce resources
• Impact of drivers of global change on distributions, risks and vulnerabilities
• Competing normative principles of fairness

Translating to Effective Policy:

• Circumstances under which policies achieve development goals
• Translating results from systems models
• Fairness of setting policy
  

POP presented a research area that looks at correlation between education and health care as a means to alleviate poverty. This is an interesting question: Can Education help solve poverty?
Africa (and South East Asia) features heavily in the worst end of the scale for mortality across the indices. It is well known (or is it?) that as your education level increases generally your income increases and hence your health. There is literature to back this up that a higher education makes more of a difference than higher income (examples: India, Indonesia) however there are a few countries where this correlation does not hold (example: Cuba). Many WHO survey graphs from 2004 were presented as a motivation but bizarrely South Africa did not show up on a single graph (Infant mortality, Mothers seeking pre-natal care etc). I asked and it seems SA is somewhere in the middle and hence frequently overlooked.

This seminar is especially interesting in light of a Poverty and Equity workshop to be held in South Africa in September. IIASA NMOs are being given an opportunity to listen to developing countries wish-lists. Why then South Africa if its not featured on the graphs? Because South Africa is going to be the nominated voice of the masses (i.e. SADC region). Does this make sense?

Something I have discovered is the POPNET newsletter published by the Population group at IIASA. The picture is from the latest newsletter that population growth to 2050 and the global education levels.

IME: Coping with Uncertainty: Concept of Robust Solutions

Tatiana Ermolieva (IME/LUC) presented what will probably be one of the most useful lectures of the YSSP for my research. Uncertainty is inherent in every decision one takes, how do you design a system that copes with it?

There are two types of systems (processes), one is the traditional natural system like physics that is governed by fixed relations, the other is human-related systems governed by old and new policies dependent on decisions of various agents/actors. From this we can distinguish knowable and unknowable (inherent: natural systems) uncertainties.

Human-related systems require a new approaches to stability (security) analysis. There are multiple reasons, the lack of observations, the expense of experimentation, observations contaminated by old policies and observations of results/impacts may come with a delay or cause irreversible change (climate change for instance)

Although its impossible to predict human-driven systems, it is possible to find robust solutions good against all uncertainties. These robust policies assist in the long-term stability of systems performance.

Popular security analysis includes aggregate analysis, least-squares analysis etc but these are not robust solutions, all are sensitive to outliers. I've appropriated the image above because it is one of the best examples I've ever seen of a mean/median/mode argument. The mean is in red, the median in green and the mode in maroon. Notice how most of the information is lost in the case of the mean. Remember: the mean is MEANINGLES! The median is more robust.
A common feature of most systems is a range of possible outcomes resulting from different scenarios or the variability (eg weather). The uncertainty is which scenario comes next?

I'm currently making my way through 'Coping with Uncertainty: Robust Solutions'. Its written by the IME group at IIASA and works through various techniques for robust systems mostly through real-world applications. The UCT library is in the process of buying a copy, it should arrive sometime in the next year :P

Science on Global Problems Area: Food & Water

IIASA has defined three new strategic research areas: Food & Water, Energy & Climate Change and Poverty & Equity.

Food & Water has been shared by a few groups within IIASA's research groups; Forestry, Ecology and Land Use Change. They reviewed current research activities and highlighted a few key challenges in this global problem area. These include:

• Widespread hunger and rising global food demands
• Agriculture needs to increase 70% globally and 100% in the developing world
• Agriculture currently uses 70% of all anthropogenic water use.
• Fishing currently feeds 1 billion people but won't be sustainable with the current overfishing
• A need for robust expansion of food and bioenergy production with sustainable regulating ecosystem functions protecting the global gene pool.
• Strategies for copying with climate change needs to be location specific.

Each group expounded on possible research avenues. Its quite fascinating what some of the proposed solutions entail. One worrying fact is that the capacity for rain-fed agriculture in Africa is decreasing by 2050, which given that 95% of the crop-land in Africa is rain-fed and contributes 35% to the GDP and 70% of the employment and 40% of the exports (according to FAO in 2000), maybe that should be a major focus on the African agenda.

PPA: Actors and Venues of Policy-Making

Today is the second Public Policy Analysis seminar by Stephen Ney where we discussed actors active or passive in policy making, these could be individuals or institutions at various levels. The diagram below shows the output of the seminar, an idea of the policy network and how they influence each other. Mine looks largely hierarchical and rigid, not as fluid and open as others. Mine also maps funding and knowledge exchanges (in red). Its interesting when writing up your research results with the anticipation that it is policy relevant one can see who to target and possibly how.

Preparatory Readings:
Atkinson and Coleman (1992), “Policy Networks, Policy Communities and the Problems of Governance”,Governance, 5(2): 154-180.
Chris Ansell (2000), “The Networked Polity: Regional Development in Western Europe”,Governance, Vol.13, No.3, pp.303-333.

Salzburg Global Seminar

The Salzburg Global Seminar is an institution that attempts to train future leaders in a diverse range of fields. The YSSP participants had the opportunity to hear Edward Mortimer talk on 'Communicating the Global Agenda', a topic he is well qualified to pontificate on. As chief speech writer for the UN Secretary-General Kofi Annan he shared his views on the global media, the pillars of the UN and provoked a quite lively lunch time discussion on how to bridge the gap between global problems (or the UN buzz word "challenges") and local politics.

The SGS is housed in yet another castle, Schloss Leopoldskron. I'm getting used to working in grandeur, going back to boring brick and concrete is going to be tough.

Not science related at all, Schloss Leopoldskron's exterior was using for all the garden shots for the Sound of Music (including the gazebo scenes and lake scenes). We had a chance to explore Salzburg in the late afternoon and spotted a few more Sound of Music sights.

After lunch and our exploration of this highly detailed schloss we entered into a very lively debate on the 'Public Understanding of Science & Public Engagement' led by Ian Brown and John Lotherington. This topic was centred around the "Climategate" incident in November 2009 that called into question the validity of climate science and the public's right to data.

Some of the chief points for and against the engagement of the public in science that emerged from this debate are summarised below (in no order):

• Will the public have more trust in scientists if they are more transparent in their methodologies and data acquisition?
• Greater engagement with the public improves the overall level of education
• Peer-review is a process with its own checks and balances and would be undermined by the public
• The allocation of funding is already decided by and large by the public through political appointments (in the case of some countries like USA but not China)
• The extra time and energy commitment needed to reply to all public criticism would hinder scientific progress
• The role of scientists is already mandated by society
• There are drawbacks to the public having access to data, it would become a popularity contest as larger institutions can release more data
• Who holds the Intellectual property rights to data produced by simulations at an government institution and the models developed there?
• Would a premature release of data create more chaos and prepared and filtered data?
• The public has a right to review any tax funded research project?
  

Any thoughts from fellow scientists? Are the public demanding too much?

IME: Multiple Criteria Analysis

Marek Makowski gave a seminar today on Multiple Criteria Analysis for the Integrated Modeling Environment (IME) group.

Integrated model analysis helps one analyse a manageable number of solutions having different trade-offs between (conflicting) criteria. There are two different types available. Descriptive (simulations) which is alternative-focused. The input is the decisions and the outcomes the goals. The other is Prescriptive (optimisation) is goal-orientated. The inputs are the goals (preferences) and the outcomes are the decisions. Where there are goals, one needs multiple criteria analysis.

IME IIASA developed a Multi-criteria Analysis tool for an energy project a few years ago. The tool is fairly easy to use and available to the public (along with some introductory and advanced papers). Our next session is a hands on workshop for using the tool.

In other news I skipped Adaptive Dynamics this morning, the PHD Comic below seems rather apt, the lecture venue is literally just down the hall.

Bioenergy: Musings of Tim Searchinger

Tim Searchinger hails from Princeton University. He gave a talk last night entitled "Bioenergy and Its Implications for Feeding the World Without Cooking the Planet" for IFF and a few YSSPers braved the deluge of rain to hear him. I knew very little going into the talk but he managed to both explain the field and the gaps of knowledge brilliantly. He has definitely been one of the best speakers I've heard this year.

Biofuels is a rather controversial topic if you read about emissions guidelines and the errors in the accounting of greenhouse gas emissions. He highlighted a few areas that the world needs to look at/study. More efficient cropland (increase yields or even look at what crop yields are...the data is terrible or unavailable), catalogue of "other" land and development of new fertilizer production with slow release nitrogen.

Final words: Don't forget the "D" in R&D

PPA: What is Public Policy Analysis

IIASA/YSSP offers students a series of seminars covering a variety of modeling and social topics. Today is the start of a six part seminar series on Public Policy Analysis by Stephan Ney. The overall objective for the course is "to introduce YSSP students to the approaches, frameworks and methods of contemporary policy analysis".

Today we delved into the realm of sociology to public policy analysis. As a scientist with no/little sociology background the question is firstly what is the definition[1], something I discovered is not actually that easy to answer.

Firstly what is POLICY? Try name some examples. You immediately get Health Care, Foreign, Education, Population etc. What about smaller scale? Gun control, Gas Emissions, Public Procurement, the list goes on. What do they all have in common? A set of rules, intentions, sequence of actions, normative and mostly government initialised. We finally agreed that Policy is an outcome/stance/process that has an underlying theory usually with with desired goals and an evaluative standard.

Some nice definitions include: "web of decisions and actions that allocate values" - Easton or "Policy is rather like the elephant - you recognise it when you see it, but you cannot easily define it" - Cunningham.

Moving on to PUBLIC, what makes a policy public? Start with what is not public like corporate. Then there are "Corporate Social Responsibility" policies. It can be argued (indeed it was) that while the corporate is pursuing some private interest it is of public interest. The consensus is that Public means it is of public interest (collective interest), usually requires some sort of oversight or government action and is defined by civil society.

Now ANALYSIS: Why analyse policies? It affects us, to improve things, efficiency, knowledge acquisition, transparency, evaluation, decision-support, effects of. You could probably name quite a few more.

[1] H. Hugh Heclo (1972), “Policy Analysis”, British Journal of Political Science, Vol.2, No.1 (Jan 1972), pp.83-108

Bring out the science heavyweights: Levien and Neureiter

Roger Levien and Norma Neureiter are both on IIASA's Endowment Board of Trustees. Roger Levien is the founder of YSSP and gave a talk on why he felt the world needed a program like YSSP. IIASA was initially founded by 12 National Member Organisations (NMOs) in 1972 as a collaboration agreement for "Applied Systems Analysis". Levien himself had been part of a graduate summer program at RAND in the USA. He describes it as a life changing experience, integrating research and policy, especially around operations research as it was then dubbed.

The funniest thing we learnt today was that the program was meant to be called "Summer Program for Young Scientists" or SPYS, which during the Cold War wasn't the most politically correct name hence YSSP.

Norman Neureiter talks about Science Diplomacy. Can you be diplomatic in Science? Neureiter believes that Science Diplomacy can bridge the cultural divide between nationals who for whatever reason disagree. His examples ranged from China, Russia, Poland and France. Each nation has research projects or exchanges with the USA that has paved the way for more political agreements. It was an interesting trip through politics, science and history from the view of a systems analyst.
Further reading for those interested:

• “The Pervasive Role of Science, Technology and Health in Foreign Policy: Imperatives for the Department of State”, Office of International Affairs, National Research Council, National Academy Press, Washington, DC, 1999.
• “New frontiers in science diplomacy”, The Royal Society, Policy Document RS1619, London, 2010
• “Science and technology in the Department of State”, by Norman P. Neureiter, Technology in Society, vol. 26 (2004), pp.303-320
• Speech on science and diplomacy by the UK Foreign Secretary, David Milliband, at the Royal Society, Jan. 12, 2010.
• Assorted articles/editorials in SCIENCE, vol. 315 (2007), p.769; vol. 305 (2004), p. 1677; vol. 326 (2009), p. 1459; etc.

2% Solution

Today we had a talk by Crawford Stanley (Buzz) Holling. Holling is an ex-director of IIASA and works in Ecological and Evolutionary Sciences and works with the Resilience Alliance. His talk was officially labeled "Broken Society, Broken Ecosystems, and Broken Planet". His talk though was primarily on the 2% solution. 98% of the time, the linear solution works. The data makes sense, but what about the 2%? Who cares? Well apparently all Systems Analysts should, this 2% falls within the realm of the exciting, the challenging and the place that allows us design new methodologies to tackle non-linear problems.

Another interesting quote came from a fellow YSSPer, "All models are bad, some are just more useful than others". Good words to live by at the moment.

Archbishop Desmond Tutu visits Michigan

Desmond Tutu is being honoured today in Michigan. Tutu will be awarded the Wallenberg Medal and deliver the 18th Wallenberg Lecture in honour of Swedish diplomat Raoul Wallenberg (U-M graduate) who saved thousands of Jews in WWII. It turns out that Tutu is the second South African to receive the medal, Helen Suzman being the first.

The press conference in the afternoon was jammed full of reporters but I managed to talk to the organisers and was allowed to sit in on part of it and be introduced to him. Apparently someone forgot to mention that Michigan is cold in October as Tutu arrived with summer suits. The organisers had to scramble to find him a coat and scarf to keep warm.

The Wallenberg Lecture itself was packed, all 3500 seats filled and an overflow hall set up. Archbishop Tutu spoke about the award going to all the people who made the end of apartheid possible. He has an incredible sense of humour that shines through on lighter moments. He encouraged the youth to dream and keep dreaming of peace and freedom globally. He critised Darfur and how rape has become a weapon of war. He finished his address with a metaphor. That an eagle should be taken and nutured and then set free from the mountain tops to soar.

Phoenix Mars Mission Talk

The seminar in AOSS this afternoon included preliminary work and results from the Phoenix Mars Mission. The speaker, Professor Nilton Renno, reviewed the mission goals of the Phoenix Mars Mission and the procedures involved in selecting a landing sight. Never trust interpolation of images was the take home message, an apparently flat surface from satellite imagery contained 10m diameter boulders. He went on to show some of his students, notably Manish Mehta whose doctoral thesis investigated the effects of the lowering rockets on the surface dust of Mars. Pretty cool work. Renno proceded to show some of the images from the Lander (inset picture) and explained some of the experiments run. It appears if you look at the images closely that there is sublimation happening on the legs of the lander with saline water settling on the mud kicked up from the landing on Mars. Apparently not everyone in the program which is shared between JPL, University of Arizona and Locklead Martin Space Systems. It is going to be interesting to see what research is published and what happens when NASA releases the data to the public in February next year.