lecture 7
Readings until now Presentations Markov, Igor L. 2014. Limits on Fundamental Limits to Computation. Nature 512 (7513) (August 13): 147 154. Sher, Stephen Loreto, Vittorio, et al. "Dynamics on expanding spaces: modeling the emergence of novelties." Creativity and Universality in Language. Springer International Publishing, 2016. 59-83. Yang, Kaicheng This week Klir, G.J. [2001]. Facets of systems Science. Springer. Chapters: 1,2,3 Optional Klir, G.J. [2001]. Facets of systems Science. Springer. Chapters: 8, 11 Lecture notes The Nature of Information Formalizing and Modeling the World Available http://canvas.iu.edu and listed at http://informatics.indiana.edu/rocha/academics/i501 Also check out Links and notes at http://sciber.blogspot.com/
The Black Box: Due October 11 th, 2017 Q1 Q3 Q2 Assignment I Due October 11 th Focus on uncovering quadrants using data collection and induction. Herbert Simon: Law discovery means only finding pattern in the data; whether the pattern will continue to hold for new data that are observed subsequently will be decided in the course of testing the law, not discovering it. The discovery process runs from particular facts to general laws that are somehow induced from them; the process of Propose a formal model or algorithm of what each quadrant is doing. testing discoveries runs from the laws to predictions of particular facts Q4 from them [...] To explain why the patterns we extract from observations frequently lead to correct predictions (when they do) requires us to face again the problem of induction, and perhaps to make some hypothesis about the uniformity of nature. But that hypothesis is neither required for, nor relevant to, the theory of discovery processes. [ ] By separating the question of pattern detection from the question of prediction, we can construct a true normative theory of discovery-a logic of discovery. Analyze, using deduction, the behavior of this algorithm.
By Erik Stolterman a possible parsing of informatics towards problem solving beyond computing into the natural and social synthesis of information technology Functionalequivalence of systems via computing and information HCID Data & Search X-Informatics or Computational X Informatics Computer Science STS, CCS, Social Informatics Security Complex Systems Data Mining Music- Health- Archaeo- Bio- Chem- Geo-
Warren Weaver classes of systems and problems organized simplicity very small number of variables Deterministic classical mathematical tools Calculus disorganized complexity very large number of variables Randomness, homogenous statistical tools organized complexity sizable number of variables which are interrelated into an organic whole study of organization whole more than sum of parts Need for new mathematical and computational tools organized complexity Weaver, W. [1948]. "Science and Complexity". American Scientist, 36(4): 536-44. http://informatics.indiana.edu/rocha
examples organized complexity Disorganized complexity Organized simplicity Organized Complexity Most relevant to problems in biology, medicine, society, and technology Randomness Complexity http://informatics.indiana.edu/rocha
From cybernetics organized complexity organized complexity study of organization whole is more than sum of parts Need for new mathematical and computational tools Massive combinatorial searches Problems that can only be tackled with computers Computer as lab
key roots systems movement Mathematics Computer Technology Systems Thinking Cybernetics Functional equivalence Communication and information Complexity Interdisciplinary outlook Bio-inspired mathematics and computing Computing/Mechanism-inspired biology and social science Kenneth Boulding Ludwig von Bertalanffy 1965: Society for the Advancement of General Systems Theory Ralph Gerard Anatol Rapoport
a science of organization across disciplines Systemhood properties of nature Robert Rosen Systems depends on a specific adjective: thinghood Systemhood: properties of arrangements of items, independent of the items Similar to setness or cardinality George Klir Organization can be studied with the mathematics of relations S = (T, R) S: a System, T: a set of things(thinghood), R: a (or set of) relation(s) (Systemhood) Examples Collections of books or music file are sets But organization of such sets are systems (alphabetically, chronologically, typologically, etc.) (complex) systems science
study of systemhood separated from thinghood (complex) systems science Study of systemhood properties Classes of isomorphic abstracted systems Search of general principles of organization Weaver s organized complexity (1948) approach Examples of subdisciplines machine learning, network science, dynamical systems theory, operations research, evolutionary systems, artificial life, artificial intelligence Works orthogonally, but tightly with classical science Interdisciplinary Systems biology, computational biology, computational social science, etc. From Klir [2001]
study of systemhood separated from thinghood (complex) systems science Study of systemhood properties Classes of isomorphic abstracted systems Search of general principles of organization Weaver s organized complexity (1948) approach Examples of subdisciplines machine learning, network science, dynamical systems theory, operations research, evolutionary systems, artificial life, artificial intelligence Works orthogonally, but tightly with classical science Interdisciplinary Systems biology, computational biology, computational social science, etc. From Klir [2001]
example of general principle of organization Barabasi-Albert Model: leads to power-law node degree distributions in networks Amaral et al: Most real networks have a cut-off distribution for high degree nodes which can be computationally modeled with vertex aging. complex networks
Informatics complex networks example of general principle of organization Barabasi-Albert Model: leads to power-law node degree distributions in networks Amaral et al: Most real networks have a cut-off distribution for high degree nodes which can be computationally modeled with vertex aging.
more formally S = (T, R) a System T = {A 1, A 2,, A n } A set (of sets) of things: thinghood Cartesian Product Set of all possible associations of elements from each set All n-tuples {A 1 A 2 A n } R: a relation (systemhood) Subset of cartesian product on T. Many relations R can be defined on the same T what is a system? x 1! x n X x 1! x n X X x x 2 1 x n x i x 1! x n X y 1! y n Y
example Equivalence classes R A B C D
example Equivalence classes R A B C D
study of systemhood separated from thinghood (complex) systems science Study of systemhood properties Classes of isomorphic abstracted systems Search of general principles of organization Weaver s organized complexity (1948) Systemhood properties preserved under suitable transformation from the set of things of one system into the set of things from the other system Divides the space of possible systems (relations) into equivalent classes Devoid of any interpretation! General systems Canonical examples of equivalence classes From Klir [2001]
Uncovering hierarchical organization From genetic interaction maps (in yeast) Jaimovich, Aet al. 2010. Modularity and directionality in genetic interaction maps. Bioinformatics 26, no. 12 (June): i228-i236.
Readings (available in OnCourse) next class Next classes Lecture Klir, G.J. and D. Elias [2003]. Architecture of Systems Problem Solving. Springer. Chapters: 1,2, 3.1, 3.2, 3.10, 4.1, 4.2 Optional: Chapters 3, 4 Coutinho, A. [2003]. "On doing science: a speech by Professor Antonio Coutinho". Economia, 4(1): 7-18, jan./jun. 2003. Knapp B, Bardenet R, Bernabeu MO, Bordas R, Bruna M, et al. (2015) Ten Simple Rules for a Successful Cross- Disciplinary Collaboration. PLoS Comput Biol 11(4): e1004214. Schwartz, M.A. [2008]. "The importance of stupidity in scientific research". Journal of Cell Science, 121: 1771.