Noise and Society: A Directed Random Walk Towards and Elementary Theory of Social Systems

Size: px

Start display at page:

Download "Noise and Society: A Directed Random Walk Towards and Elementary Theory of Social Systems"

Neal Blake
5 years ago
Views:

1 Noise and Society: A Directed Random Walk Towards and Elementary Theory of Social Systems Dirk Helbing (ETH Zurich) with Thomas Chadefaux, Karsten Donnay, Andreas Flache, Anders Johansson, Stefan Lämmer, Sergi Lozano, Michael Mäs, Mehdi Moussaid, Matjaz Perc, György Szabo, Attila Szolnoki, Guy Theraulaz, and Wenjian Yu 9/5/2008 Sunday, November 28, 20101

Sociology - The Queen of Sciences? Auguste Comte (1798-1857) is often called the father of sociology.

2 Sociology - The Queen of Sciences? Auguste Comte ( ) is often called the father of sociology. He proposed a rational ( positivistic ) approach to the study of society, based on observation and experiment. In the beginning, he called his approach social physics, but later he used the term sociology (meaning knowledge of society). Auguste Comte considered sociology to be the queen of sciences. Comparing, for example, sociology with biology and physics, the systems it deals with are the most complex ones.

3 General Problem in the Social Sciences We cannot understand society because of the noise (in the data)

4 The Challenge of Simplification and Abstraction

5 A Note on Simple Models Geocentric Picture: Epicycles around the Earth Heliocentric Picture: Elliptical paths around the sun

6 Can We Understand a (Social) System from Elementary Processes?

7 Model Ingredients: Elementary Properties of Individuals Birth, death, and reproduction Individuals need resources (e.g. eat and drink) Competition, fighting ability Tool-making ability, possibility to grow food, hunt etc. Perception Curiosity, exploration behavior, ability for innovation Emotions (Memory) Mobility and carrying capacity Communication Learning and teaching ability Possibility of trading and exchange

8 Scientific Puzzle Why do we have global opinion diversity despite local convergence?

9 Stylized Facts of Collective Opinion Formation Individuals are influenced by other people s stated opinions Collective opinion formation creates a tendency of convergence However, consensus does not always happen If consensus occurs, this creates confidence in the correctness of the opinion People tend to form groups, but groups may also split up As a result, the typical observation is a pluralistic society Emergent groups are characterized by similar opinions and/or behaviors ( homophily )

10 Consensus and Overconfidence Undermine the Wisdom of Crowds

11 Limitations of the Bounded Confidence Model of Opinion Formation Joint work with Michael Maes and Andreas Flache, in PLoS Computational Biology

12 Collective Opinion Dynamics According to the Durkheimian Model Joint work with Michael Maes and Andreas Flache, in PLoS Computational Biology

13 Pluralism Exists Between Anomie and Opinion Monoculture

14 Scientific Puzzle Why and when do selfish individuals cooperate in social dilemma situations?

15 The Importance of Social Cooperation Social cooperation is expected to break down, if selfish free-riding generates better payoffs The generation of public goods requires to overcome this tragedy of the commons Everything from public services and infrastructures over protecting our environment up to social benefit system and language and culture requires cooperative contributions Without social cooperation, societies would not exist We would have a war of everybody against everybody else ( homo hominis lupus ) The challenge is to understand how to overcome the tragedy of the commons Taxes and a state enforcing cooperation are one solution. Do others exist?

16 Some Relevant Stylized Facts Individuals like to agglomerate (form cities, groups, etc.) Individuals with different behavioral strategies tend to segregate (--> see also Schelling s model) Levels of cooperation in the prisoner s dilemma and in public goods games are higher than expected Cooperation tends to break down ( tragedy of the commons ), but may thrive in closely interacting groups and if people can leave bad environments and choose more favorable ones Individual behaviors are partially determined by the social environment they are contributing to (--> norms) Social environments persist much longer than an average individual contributes to it (--> social institutions) However, social systems adapt continuously

17 Imitation and Success-Driven Motion, Separately and Together P = 0 R = 1 S = 0 T = 1.4 blue = C red = D imitation only migration only imitation & migration

18 The Breakdown and Outbreak of Cooperation Red, yellow: defectors (cheaters) Blue, green: cooperators

19 Does Globalization Endanger Social Cooperation? Local interaction, local migration Global interaction, local migration Global interaction, global migration

20 How the Rich-Gets-Richer-Effect Can Support Cooperation How much inequality is needed for the cooperation-enhancing effect?

21 Social Order: How Important is the Protection of Private Property?

22 The Positive Effects of Moderate Noise Noise can overcome metastable, but suboptimal configurations

23 Global Warming and the Spreading of Costly Punishment Imagine that cooperators (C) correspond to countries trying to meet the CO 2 emission standards of the Koyto protocol, and moralists (M) to cooperative countries that additionally enforce the standards by international pressure (e.g. embargoes). Defectors (D) would correspond to those countries ignoring the Kyoto protocol, and immoralists (I) to countries failing to meet the Kyoto standards, but nevertheless imposing pressure on other countries to fulfil them. For well-mixed interactions, defectors will be the winners of the evolutionary competition among the strategies, i.e. all countries would finally fail to meet the emission standards ( tragedy of the commons ). The reason is that cooperators ( second-order freeriders ) spread at the cost of moralists, while requiring them for their own survival. (See the work of Milinski et al.!)

24 How Second-Order Free-Riders Are Eliminated+Punishment Spreads D = Defectors (Free-Riders), M = Moralists, I=Immoralists C = Non-punishing Cooperators (Second-Order Free-Riders)

25 The Unholy Symbiosis of Moralists and Immoralists D = Defectors (Free-Riders), M = Moralists, I=Immoralists C = Non-punishing Cooperators (Second-Order Free-Riders)

26 Overcoming the Tragedy of the Commons by Spatial Interactions

27 Lucifer s Positive Side Effects

28 Conclusion We cannot understand society without considering noise

It turns out, however, that our current knowledge of society is too limited to efficiently tackle the global challenges of

29 The Knowledge Accelerator Unleashing the Power of Information for a Sustainable Future Dirk Helbing, with the support of >300 scientists from all over the World We have explored the universe, and have sent men to the moon. It turns out, however, that our current knowledge of society is too limited to efficiently tackle the global challenges of humanity in the 21 st century. Thus, it s time to pay attention to our Earth and create an ICT Flagship to explore social life and everything it relates to.

Challenges Humanity is Facing in the 21st Century Lee C. Bollinger, president of Columbia University, formulated the issue as follows: The forces affecting societies around the world.

These questions call for the kinds of analyses and understandings that academic institutions are uniquely capable of providing. Too many policy failures are fundamentally failures of knowledge. 1.

30 Challenges Humanity is Facing in the 21st Century Lee C. Bollinger, president of Columbia University, formulated the issue as follows: The forces affecting societies around the world... are powerful and novel. The spread of global market systems... are... reshaping our world..., raising profound questions. These questions call for the kinds of analyses and understandings that academic institutions are uniquely capable of providing. Too many policy failures are fundamentally failures of knowledge. 1. Financial and economic crisis 2. Debts and inflation 3. Stability of the European Union 4. Corruption 5. Organized crime, hooliganism 6. Extremism, terrorism, war 7. Epidemics (SARS, H1N1 pandemic) 8. Security and cyber risks 9. Migration and integration 10. Environmental change

31 The Top 10 Socio-Economic Problems and their Reasons Problems: 1. Demographic change, migration 2. Financial and economic stability 3. Social, economic and political inclusion, integration 4. Public health 5. Balance of power and conflict 6. Corruption and crime 7. Collective social behavior 8. Institutional design 9. Sustainable use of resources 10. Reliability of critical infrastructures Reasons: 1. Interdependency, interconnectivity 2. Socio-economic, ecological, and technological complexity 3. Self-organization, emergence, chaos 4. Limits of predictability and control Cascade failures/ avalanche effects: Epidemic spreading, congestion spreading, failure of interbank market, breakdown of former GDR 5. Lack of quantitative models 6. (Due to) Lack of data 7. Lack of computational power 8. Lack of systemic predictions 9. Lack of tested alternatives 10. Systemic risks This is about to change!

by massive cascading effects ( domino effects, avalanche effects ) Triggered by overcritical perturbation or coincidence of

32 Cascade Spreading and Systemic Crises Network interactions are ubiquitous Feedback loops, circuli vitiosi Unwanted side effects Systemic malfunctions, whenever the system state changes beyond a critical threshold ( tipping point ) Often caused by massive cascading effects ( domino effects, avalanche effects ) Triggered by overcritical perturbation or coincidence of failures Examples: Epidemic spreading, failure of interbank market, congestion spreading, blackout of electrical power system

33 The Need of A Knowledge Accelerator SOCIETY NATURE We need to create a techno-socio-economicecological knowledge accelerator - a kind of multidisciplinary Apollo project that uses current and future ICT developments to address the challenges of humanity, involving natural scientists and engineers

34 New ICT for Socio-Economic-Ecological Reality Mining + Simulation Sensor networks Remote sensing Web2.0 Internet Second life Satellites Telecommunication Prediction markets GPS Social networks

35 Overview of FuturICT s Living Earth Visualator Concept Data collector Testing of alternative solutions Decision arena European-scale, multi-disciplinary effort is needed! Crisis observatory Political decision-making

36 Lord Robert May of Oxford: Your letter to Soros puts the case admirably well, and I believe he may well be interested in such an initiative.

37 Decentralized Concept of Self-Organized Traffic Light Control Inspiration: Selforganized oscillations at bottlenecks Measurement input Optimal compromise between coordination and local flexibility Published in JSTAT (2008)

38 Self-Organized Control of a Single Intersection Measurement input Published in JSTAT (2008)

39 Self-Control of Urban Traffic: Environmental-Friendly without Pain

40 Superexponential Bubbles or Crashes Instead of Stable Growth Credits to Didier Sornette Black Monday in 1987 was a 35 sigma event!)

41 The Need for Bubble Control Mode of heat transfer depends on energy input: 1. diffusion 2. convection rolls 3. bubbles Does liquidity play a similar role?

42 Thank you for your interest! Any questions?

Contents Modeling of Socio-Economic Systems Agent-Based Modeling

Contents Modeling of Socio-Economic Systems Agent-Based Modeling Contents 1 Modeling of Socio-Economic Systems... 1 1.1 Introduction... 1 1.2 Particular Difficulties of Modeling Socio-Economic Systems... 2 1.3 Modeling Approaches... 4 1.3.1 Qualitative Descriptions...