Artificial Life (ALife) Course - Intro Slides PDF

philosophy in practice Vida Artificial / Artificial Life (ALife) (MEI) & Modelos de Computação (MCCog) Luı́s Correia DI, Lasige – FCUL philosophy in practice in fact it shoul...

philosophy in practice Vida Artificial / Artificial Life (ALife) (MEI) & Modelos de Computação (MCCog) Luı́s Correia DI, Lasige – FCUL philosophy in practice in fact it should read Computational Artificial Life since we do not cover “wet” ALife  what is ALife? philosophy in practice Artificial life is a field of study devoted to under- standing life by attempting to abstract the fundamental principles underlying biological phenomena, and recre- ating these dynamics in other media – such as com- puters – making them accessible to new kinds of ex- perimental manipulation and testing Langton (1991) Artificial Life 2, Preface life on Earth: life as we know it artificial life: life as it could be  ALife in short philosophy in practice Condensed description Studies synthetic behaviour similar to biologic organisms, in computers, mobile robots and more  ALife two families of goals philosophy in practice using artificial life techniques... modelling what can we learn about the living? engineering how can we obtain better solutions to known problems? in this course, we are going to cover some of the techniques, algorithms and models that are used to study questions like these  an illustrative example philosophy in practice morphology and neural control under evolution some beings also have Karl Sims ’blokies’ (1994) sensory inputs [external video] evolutionary pressure to move to capture problem in development phase: ’self-propelling beings’  RQs big questions! Artificial life - RQs philosophy in practice evolution of complexity – is there a reason for the arrow of complexity? origins of life – understanding pre-biotic systems (template replicators, etc.) ETI* major transitions in evolution prokaryotes to eukaryotes, solitary to eusocial colony living, single to multi-cellular life * evolutionary transitions in individuality how does cultural evolution influence biological evolution, and vice versa? what are the fundamental requirements for intelligence? Bedau et al (2000) Artificial Life 6:363  course logistics I philosophy in practice Luı́s Correia (T, TP11) Nuno Garcia (TP12) [email protected] [email protected] course email: [email protected]  course logistics II philosophy 2h T + 1.5h TP [TP11, TP12] we need to balance in practice students in the 2 TPs assessment project [70%] groups of 2 2 pg. proposals by 24/Oct (by email); pres TPs 27/Oct complete project by 23/Jan, presentation 26/Jan flash test [20%], 20 questions, 20 mins in last T class 5x 2pg-reports from TP exercises [10%] contact [email protected]  course logistics III philosophy in practice project free theme on any of course topics, or related Vida Artificial implementation project Modelos de Computação implementation project or synthesis report  your turn philosophy background in practice study and research interests what you want to learn complementary skills  course overview philosophy in practice historical notes dynamical systems cellular automata evolutionary algorithms neural networks swarm models self-organisation mobile robots (as artificial beings)  references I introduction to ALife philosophy Christoph Adami, “Introduction to Artificial Life”, in practice Springer Telos, 1998 Cristopher Langton, “Introduction”, pp. 1-47, in Cristopher Langton, ed., Artificial Life, Proceedings of the 1987 Workshop, Los Alamos, Addison-Wesley, 1989 Cristopher Langton, “Introduction”, pp. 3-23, in Langton, Taylor, Farmer, Rasmussen, eds., Artificial Life II, Addison-Wesley, 1992 Christopher Langton ed., Artificial Life: An Overview, 1997 Artificial Life Online – https://alife.org/  references II introduction to ALife Steven Levy, Artificial Life: The Quest for a New Creation, philosophy in practice Panteon Books, 1992 Claus Emmeche, The Garden in the Machine, Princeton University Press, 1991 Erwin Schrödinger, O que é a Vida?, Fragmentos, 1989 (original de 1943) Heins Pagels, Os Sonhos da Razão, Gradiva, Gradiva, 90 (original de 88) Edgar Morin, Para o Pensamento Complexo, in Ciência com Consciência, Publicações Europa- América, 1994 (original de 1990) John Horgan, From Complexity to Perplexity, Scientific American, Jun. 1995, pp. 74 - 79.  ALife context philosophy in practice  characteristics living systems (& ALife) philosophy in practice autonomy multiple similar components self-organisation local control reproduction evolution emergent properties  ALife context philosophy in practice Life forms vs. environments LIFE examples natural artificial natural Bio-organisms Artificial beings Cellular Automata WORLD artificial Virtual reality Virus Sofbots  Parallels philosophy in practice Some say As Artificial Intelligence is to intelligence, so Artificial Life is to life...  ALife Pre-history... philosophy in practice Vaucanson’s duck (∼1730)  Near ancestors philosophy in practice Artificial neuron model (∼1940) Turing’s morphogenesis (∼1950) von Neumann’s cellular automata (∼1950) Grey Walter turtles (∼1950)  The turtles by Grey Walter philosophy in practice Machina Speculatrix - Elmer e Elsie Very simple behaviour If collision with an obstacle go away Approach light sources, except if light source too strong go away however, resulting interaction with environment is complex  Turtles in action philosophy in practice © Burden Neurological Institute  The beginning of ALife philosophy in practice I Workshop on ALIfe in 1987 To join researchers in several scientific domains, apparently with common interest in self-organised systems Organised in Los Alamos, by Christopher Langton (Santa-Fe Institute)... and continuation philosophy in practice Currently Conferences A-Life Conference ECAL - European Conference on Artificial Life SAB - Simulation of Adaptive Behavior Journals Artificial Life Adaptive Behavior Evolutionary Computation Transactions on Systems Man & Cybernetics Transactions on Evolutionary Computation Autonomous Robots  ALife vs. AI philosophy in practice appears as opposing ALife ←→ AI behaviour ←→ solution dynamic evolution ←→ final state essentially parallel ←→ essentially sequential action ←→ reasoning  Epistemology I/III philosophy in practice what is life? living organism has: Reproduction Evolution Phylogenetic - species Ontogenetic - integration of interactions with environment (learning) - nervous, immune and endocrine systems Epigenetic - individual growing, morphogenesis etc.  Epistemology II/III philosophy in practice Is life a property of form or a property of mater? Von Neumann’s self-reproduction model preceded discovery of biological reproduction It is difficult (impossible?) to draw a clear line between life and non-life, but...  epistemology III/III simplifying philosophy Pier Luigi Luisi: in practice All that is alive is constituted by cells with properties of: self-maintenance self-generation by internal activity of the organism, consuming: Energy/Nutrients...  under life philosophy Self-organised systems in practice Composed of: simple components with emergent properties showing at the system level, but... nonexistent at the component level wider than ALife Ex: Temperature, pressure, turbulence, meteorologic phenomena, economy, society,...  Self-organisation defining principles philosophy in practice 1 No external control in nature all SO is embodied ⇒ lessons to take to artificial SO  Self-organisation defining principles philosophy in practice 1 No external control 2 Increase in order in nature all SO is embodied ⇒ lessons to take to artificial SO  Self-organisation defining principles philosophy in practice 1 No external control 2 Increase in order 3 Adaptability in nature all SO is embodied ⇒ lessons to take to artificial SO  Self-organisation defining principles philosophy in practice 1 No external control 2 Increase in order 3 Adaptability 4 Interaction in nature all SO is embodied ⇒ lessons to take to artificial SO  Self-organisation defining principles philosophy in practice 1 No external control 2 Increase in order 3 Adaptability 4 Interaction 5 Asynchronism in nature all SO is embodied ⇒ lessons to take to artificial SO  by the way of emergence the observer philosophy in practice in most cases: what emerges is in the eye of the beholder situation of syntactic emergence [Cariani] the system in itself did not acquire new capabilities the model of the system does not change to encompass the observed behaviour  by the way of emergence the system philosophy in practice What emerges is in the system does it exist in computational systems? situation of semantic emergence [Cariani] the system acquires new capabilities we must change the system’s model to explain the observed behaviour  by the way of emergence the system philosophy in practice What emerges is in the system does it exist in computational systems? situation of semantic emergence [Cariani] the system acquires new capabilities we must change the system’s model to explain the observed behaviour In a computational system we do not have more than a Turing machine = finite state machine ⇒ non emergent!  ways to approach ALife philosophy in practice WEAK position ALife as a family of models to solve different types of problems interest in knowing what is done in other areas and to apply those ideas STRONG position ALife as a unifying paradigm underlying different life forms interest in finding fundamental parameters (models), common to all forms  The other side of the mirror... critics philosophy critic to Unifying Theory or Unifying Principle of ALife and in practice self-organised systems Horgan, 95 ALife is the philosophical heir of AI. As such, it generated more portentous rhetoric than tangible results Most important finding of ALife: It is very difficult to do science in complex systems  Compared Granularity life, natural vs. artificial philosophy in practice Nature Artificial sub-atomic particles - atoms e molecules - carbon technology silicon technology - materials technology physico-chemical interactions electric interaction organic molecules integrated circuits cells (neurons) software/hardware organs software integrated organism electronics + mechanics  important relations philosophy in practice Evolutionary Algorithms Physics Complex Artificial Dynamic S. Immune S. Life Synthetic Cellular Chemistry Automata Ethology  applications philosophy in practice modelling of biological systems modelling of socio–economic systems synthesis of artificial organisms in simulated environments (Tierra, iterated prisoner’s dilemma) and in real environments (mobile robots) solving of complex problmes (alike grid computing)... wet ALife is a wide different(?) area...