0\F0$9$k%^%k%A!&%(!<%8% %s%h!&%7%9%f% %s%h!&%7%9%f% $NF0:n%"%k%4%j%:% $N<+8J:FJT$K8~$1$

Size: px
Start display at page:

Download "0\F0$9$k%^%k%A!&%(!<%8% %s%h!&%7%9%f% %s%h!&%7%9%f% $NF0:n%"%k%4%j%:% $N<+8J:FJT$K8~$1$"

Transcription

1 !LRESEARCH!M Generating the Explanation of the Complex Behaviors toward Self-Reorganization of Mobile Multi-Agent Systems Koichi KURUMATANI Mari NAKAMURA In order to deal with unexpected or illegal behavior of multi-agent systems, we have to reorganize the system's behavior by redening or re-planning each agent's behavior. In that case, the underlying mathematical model connecting the target system's behavior and each agent's behavior is indispensable. In this paper, we present a method for generating causal relations among the parameters which describe the target multi-agent system, consisting of arithmetic and dierential relations of explicitly dened parameters of agents and implicitly existing parameters embedded in the target system. The task consists of three components, 1) micro-macro translator which reasons about macrobehavior rules from the given micro-behavior of each agent, 2) causal network constructor with qualitative region grammar, and, 3) macro-behavior qualitative simulator. We took an example for the method from the foraging behavior of ant colonies, which is a typical mobile multi-agent system with a local communication method by chemical pheromone. x 1 Introduction Methods for automatically redening mobile agents' behavior-plans, and as the eect, for reorganizing the mobile multi-agent system's behavior are indispensable to eciently achieve the system's target. For instance, an automated cargo-control system which transports baggage from airport counters to airplanes and vice versa should have global goals of enlarging the transport ratio (pieces of baggage per hour) and shortening the total path length of cargo-carriers (reducing the energy consumption), under the constraints of the restriction of the area where the carriers move, collision avoidance, carrying-time limit (cargos should reach the plane by departure time), and so on. Since events such as plane delays, troubles in cargocarriers, and weather changes, unexpectedly occur in the system, the system cannot plan the exact schedule beforehand and should always observe the system's behavior and reorganize the system's behavior by replanning each agent's behavior. The scientic questions about the complex behaviors of the multi-agent systems arise in connection to the emergent property 1) 2) of the systems' behaviors, i.e., the systems' behaviors can be very complex in spite of the simple behavior of each agent. Although we have no way to entirely control such properties, we can understand and explain the behaviors once we get the underlying mathematical model by observing the system's behavior and verifying the assumptions about the tar- KEYWORDS: mobile multi-agent system, self-reorganization, explanation generation, qualitative region grammar, causal network, planning, articial life {(1){

2 2 Bulletin of the Electrotechnical Laboratory Vol. Vol No.No (1995) get system. In this paper, we propose a method for generating causal relations among the parameters which describe the target multi-agent system for re-planning each agent's behavior. By causal relations we mean the arithmetic and dierential relations of parameters in qualitative manner, which changes qualitatively according to the system status. All parameters which need to describe the target system is not given explicitly. There exists implicit parameters embedded in the target system, which must be salvaged. The method consists of three components, 1) micromacro translator which reasons about macro-behavior rules from the given micro-behavior of each agent, 2) causal network constructor with qualitative region grammar, and, 3) macro-behavior qualitative simulator. In the following section, we illustrates the example selected for the method, the foraging behavior of ant colonies with the communication method by chemical pheromone. x 2 Problem 2.1 Modeling the Foraging Behavior of Ant Colonies We selected an ant colony as an example for the causal network construction because it is a typical example of mobile multi-agent systems. It has macrogoals in the colony level which should be achieved by the cooperated behaviors of micro-agents (ants). Microagents have a local communication method by chemical pheromone, but the colony itself has no global communication methods, e.g., telecommunication, global map of environment, and so on. Because the lack of global communication methods, it is dicult for the target system to have centralized controlling mechanisms (the headquarters of ant colonies), and thus the target system should achieve macro-goals by coordinating or tuning micro-agents' behaviors. The nal goal of our research is to nd a way to coordinate or tune the micro-agents' behavior by 1) observing the macro-behavior of the target system, 2) constructing the underlying mathematical model (causal relations in the target system), and 3) redening the micro-behavior of each agent. In this paper, we describe the method for constructing the causal relations of the target system and generating the causal network. The foraging behavior of ant colony is an organized behavior of the ant society, and is a typical example of complex behaviors of biological multi-agent systems. Although the behavior (algorithm) of each ant is quite simple one with a communication method by chemical pheromone, the colony shows complex foraging behaviors which maximize the bait transport ratio and minimize the risk caused by environmental disturbance, i.e., climate, food competition, and so on 3). In this paper, the model of the foraging behavior of each ant is assumed as follows (they are described as an automaton in the reasoning system) (Fig.1) y1. [The foraging behavior of ant colonies] 1. At any time, an ant is in a mode, search, attracted, trace, or transport. 2. Search is the default mode. The ant walks randomly in search mode. 3. When an ant in any mode nds a bait-site, it tunes into transport mode, where it carries a bit of bait to the colony's nest. Bait can exist at several baitsites. Ants in transport mode secrete the recruitment pheromone on the line of the transportation which becomes the pheromone trail. The ant in transport mode becomes in search mode when they reaches the nest. 4. (The pheromone trail evaporates and diuses, which makes pheromone atmosphere.) 5. When an ant in search mode comes across pheromone atmosphere, it turns into attracted mode and is induced by the pheromone to the higher pheromone density. If it will nd pheromone y1 The model is not entirely faithful to the real ant behavior. It represents common aspects of many kinds of ant colonies. {(2){

3 0\F0$9$k%^%k%A!&%(!<%8% %s%h!&%7%9%f% %s%h!&%7%9%f% $NF0:n%"%k%4%j%:% $N<+8J:FJT$K8~$1$ trail, it traces the trail to a bait-site (in trace mode). Otherwise it continues to walk in the pheromone atmosphere. This simple algorithm of the ant (agent) implements the foraging behavior of the ant colony (multi-agent system), i.e., 1. to nd the site(s) of bait, 2. to mobilize ants in the colony in order to carry bait on a large scale, therefore, 3. to maximize bait transport ratio per time. Notice that this simple agent's algorithm (and the colony's strategy) can be easily inuenced by the environmental disturbance. When more than one bait sites exist and all the ants in the colony gather to a bait place simultaneously, enemy animals can easily attack the colony, or the colony might miss the other bait places which have more amount of bait than the current attacked bait place. Nest Search Mode Ant Transport Mode Ant Attracted Mode Ant Pheromone Trail Area of Pheromone Inducement (a) Behavior of each ant. Bait Place Trace Mode Ant Pheromone_density > threshold Area of Pheromone Inducement Diffusion Evaporation Active Trail Trail_magnitude > threshold Decreased Trail Trail_magnitude < threshold (b) Evaporation and diusion of pheromone. Fig.1. The foraging behavior system of ant colonies. 2.2 The Results of Numerical Simulations We have carried out the numerical simulations of the target system for several sets of parameters. A part of the results of the simulations is shown in Fig.2. In these simulations, there exists a nest in the center of the environment, and there exist eight bait places distributed in the same distance from the nest place. The dierence of the condition is only the number of ants in the colony. In Fig.2(a), there exist 60 ants searching 8 bait places, and some ants actually nd a bait place and generate a pheromone tail between the bait place and the nest. Since the trail evaporates quicker than the other ants gather to the trail, the continuous growth of the trail and the continuous large-scale transport are not achieved. In Fig.2(b), enough 600 ants gather to the trail, and the large-scale transport is achieved. The results tell that enough number of agents should exist in the system in order to conquer the time-delay between the gathering speed of ants and the evaporation speed of the pheromone, which can be explained by the causal relations embedded in the target system. In Fig.2(b), almost all ants gathers to the bait number 5, and the colony `forget' the other bait places. Actually the same phenomena that all the ants gathers to only one bait place is observed under almost all parameters whenever the large-scale transport is realized. The reason is that there exists a positive feedback among the dierential relations of the system parameters, i.e., the number of ants gathering to a bait, the amount of secreted pheromone by the ants, and the amount of evaporated pheromone atmosphere which attracts ants. Our nal target of this research is to redene or replan each agent's behavior to eciently achieve the system's goal, by propagating qualitative values on such causal networks. In the following section, we discuss a method for generating the causal networks, i.e., qualitative region grammar, causal network constructor, micro-macro translator. {(3){

4 4 Bulletin of the Electrotechnical Laboratory Vol. Vol No.No (1995) Search mode Attracted mode Trace mode Transport mode Bait 1 4 Bait 1 Nest Trail Nest Area of Pheromone Inducement 6 7 (a) Total 60 ants. (b) Total 600 ants. Fig.2.1. The map of ants after 1000 steps (a) Total 60 ants. t= ants 8 (b) Total 600 ants. Fig.2.2. The number of ants which reached the bate places. t= ants 100 % 0 t=0 (a) Total 60 ants. t=1000 Attracted Trace 100 Search % Transport 0 t=0 (b) Total 600 ants. t=1000 Fig.2.3. The percentage of ants in each mode. Fig.2. The results of the numerical simulation of the foraging behavior for 60 and 600 ants. {(4){

5 0\F0$9$k%^%k%A!&%(!<%8% %s%h!&%7%9%f% %s%h!&%7%9%f% $NF0:n%"%k%4%j%:% $N<+8J:FJT$K8~$1$ x 3 Generating Causal Networks The simulation of the complex behaviors of multiagent systems have been already studied 4), but they lack the underlying mathematical models. Statistical analysis for system behaviors with a xed simple behavior of agent 5) can describe the statistic relationship and distribution between agent's behavior and system's behavior, but they lack of the causal relations, and the explanation and re-planning ability are impossible. Self-Explanatory Simulation 6) generates explanations by causal relations, which is limited to simple physical systems (a ying ball, a pulley). Our approach is to prepare symbolic representations in which causal relations among parameters, spatial extents, and attributes of the target system are described, and to propagate qualitative values on the derived causal networks for re-planning of the agent's behavior. The method has the advantages 1) that spatial extent can be represented and reasoned about symbolically, therefore partial dierential equations (e.g., diusion) can be solved in qualitative (symbolic) manner, 2) that translation between micro and macro behaviors is possible, and 3) that the generated causal networks can be used to redene or re-plan the target multi-agent system, e.g., repressing the exponential amplication in the foraging behavior which is found in a causal network for ant colonies. 3.1 Qualitative Region Grammar By qualitative region we represent the spatial extent describing the attributes of the target system and their changes, e.g., how the pheromone tail or atmosphere spreads, or how ants in search mode are distributed. Mathematically saying, when a function f(x; y) and a meaningful value (landmark) l are given, a qualitative region r = region(f, l) denotes a closed line in x-y plane which is represented by f(x; y) = l: Usually we use the notation of r = region(f, l, +) which represents the closed region f(x; y) > l in the x-y plane (Fig.3) y2. Since our aim is not to represent such regions precisely (it can be done by numerical simulations), we represent their characteristic properties symbolically and to reason about behaviors of the target systems with them. landmark value O Fig.3. z (f(x,y)) C1 y C2 f(x,y) Denition of qualitative regions. We represent the characteristic properties of regions with predicates: static spatial relations between two regions identical(ra, rb), isolated(ra, rb), intersect(ra, rb), include(ra, rb), broader(ra, rb) dynamic changes of a region (t 1 > t 2 ) expand(r, t1, t2), shrink(r, t1, t2), no_change(r, t1, t2) expanding(r, t1), shrinking(r, t1), appear(r, t1), disappear(r, t1) some basic inference rules (independent from the target system) no_change(r, T1, T2) :- constant(r). expand(r, t1, t2) :- broader(region(r, t1), region(r, t2)). expand(r, t1, t2) :- include(region(r, t1), region(r, t2)). y2 If there exist more than one such regions, region(f, l, +) denotes separated regions r1, r2,..., rn. x {(5){

6 6 Bulletin of the Electrotechnical Laboratory Vol. Vol No.No (1995) no_change(r, t1, t2) :- identical(region(r, t1), region(r, t2)). expanding(r, t1) :- expand(r, t1 + delta_t, t1). 3.2 Macro-Behavior Rules A causal network is a graph representing the causal relations among the parameters, spatial extent, and attributes of the target system, on which qualitative properties are propagated. The nodes of the graph represent parameters of the macro behavior, i.e., the behavior of the target system rather than each agent (the behavior of each agent is called `micro behavior'), and the arcs are labeled by the relation of adjacent nodes. For instance, in a situation where ants in search mode come across pheromone atmosphere and they turn into attracted mode, the transition ratio (frequency) of the ants from search mode to attracted mode is qualitatively proportional to the density (population) of the ants in search mode, and also to the area of pheromone atmosphere, i.e., y3 M0+(s_a_trans, search_ant_pop). M0+(s_a_trans, area(pheromone_atmos)). Such a macro-behavior reasoning rule is used to generate the causal networks, and is written in the form of predicate: transit(search, attracted) :- region(s, search_region), region(p, pheromone_atmos_region), intersect(s, P), transit_s_a(p). transit_s_a(phero) :- /* new region */ assertz( region(attracted_zero, attracted_region)), assertz(include(phero, attracted_zero)), /* transition frequency */ assertz( trans_freq(s_a_trans, search, attracted)), assertz( y3 M0+(a,b) means that a is qualitatively proportional to b, is positive, and a = 0 when b = 0. m_zero_plus(s_a_trans, search_ant_pop)), assertz( m_zero_plus(s_a_trans, area(phero))). The macro-behavior reasoning rules are generated by the micro-macro translator 7) at the beginning of the reasoning process. The micro-macro translator receives an automaton which describes the agent behavior (in section 2.1), and generates macro-behavior reasoning rules which are used to construct the causal networks throughout the reasoning process. 3.3 Micro-Macro Translator The reasoner receives 1) agent's behavior as an automaton, 2) descriptions of environment (e.g., how the evaporation of pheromone is represented in qualitative region grammar), and 3) the initial state of the target system as each agent's state. First of all, the reasoner invokes the micro-macro translator in order to generate macro-behavior reasoning rules. The translator is a production system which receives an automaton of the micro-behavior description as a state transition graph, and translates the graph to macro-behavior rules by applying production rules which are independent from the target system. The independence of the rules from the target system is not surprising, because our targets are limited to `mobile' multi-agent systems, therefore the micro-macro translation can exploit the metric of Euclidean spaces. For instance, the positions of mobile agents, which are originally represented as parameters in micro-behavior level, can be easily translated to qualitative regions in macro-behavior vocabulary, since the metric of the twodimensional plane prepares a simple mapping between position (point) and spatial extent (region). When we apply our approach to other kinds of multi-agent systems which do not originally have the metric, the micromacro translation could become target-dependent, but even in such cases there is a possibility that the environment of the target system is metrizable. In the current implementation, we assume that mo- {(6){

7 0\F0$9$k%^%k%A!&%(!<%8% %s%h!&%7%9%f% %s%h!&%7%9%f% $NF0:n%"%k%4%j%:% $N<+8J:FJT$K8~$1$ bile agents stay in two-dimensional plane. When they moves in three-dimensional spaces, we need to modify the qualitative region grammar which will have more complex inference rules. in Fig.2(b), where enough numbers of ant are attracted by pheromone, and will trace the trail and transport the bait with secreting recruitment pheromone. 3.4 Collecting Qualitative Relations for Constructing Causal Networks The macro-behavior reasoning rules generated by the micro-macro translator are the collection of the declaration \how the target system behaves in a certain situation" described in macro-level. We reason about the target system's behavior in the following process. population(search_1) M0+ Q- trans_rate(search_1, attracted_1) Q+ population(attracted_1) M0+ Q- trans_rate(attracted_1, trace_1) Q+ population(trace_1) total_amount(trail_1) Q- evaporation_rate = + (const) Q+ total_amount(phero_1) Q- M0+ diffusion_rate = + 1. Set the initial state to the current state. 2. Apply the macro-behavior reasoning rules to the current state, and determine the rule to become active. M0+ Q- Q+ M0+ trans_rate(trace_1, transport_1) area(phero_1) Q+ population(transport_1) Fig.4. A generated causal network. 3. If there is no activated rule, stop. 4. Collect the qualitative (causal) relations of parameters in activated rules, i.e., generating the causal network corresponding to the current state. 5. Examine the possibility for each parameter to cross the landmark which denes the corresponding qualitative region (spatial version of limit analysis). 6. Collect the fragments of the next state in the activated rules, and set it to the current state. 7. Go to 2. This process resembles one in Qualitative Process Theory 8) 9). The dierence is that 1) the macrobehavior reasoning rule, which corresponds to individual view and process in Qualitative Process Theory, is generated from micro-behavior of agents by the translator, 2) spatial extents such as pheromone diusion can be represented and reasoned about by qualitative regions. An example of the generated causal networks is shown in Fig.4 y4. This network represents the state x 4 Explanation Generation An application of the causal network is to generate the explanation of the system's behavior by propagating the qualitative value in the network. The reasoner computes the directions of change of parameters (increasing, decreasing, or stable) and generates the explanation of the state. In the case of the transition from search mode to attracted mode, the following explanation is generated. transition from search mode to attracted mode is proportional to the population of search ants. the number of search ants is positive. the transition is positive.... Such an explanation is generated for each state until the reasoner nds no more possibility of state transitions of the target system. As a result, explanations along the state transitions of the target system are obtained. y4 represented in predicate form in the system. the area of pheromone atmosphere expands. {(7){

8 8 Bulletin of the Electrotechnical Laboratory Vol. Vol No.No (1995) transition from search to attracted is positive. transition from attracted to trace is positive. transition from trace to transport is positive. 1) Andrew M. Assad and Norman H. Packard: Emergent colonization in an articial ecology, In Fran- the area of pheromone atmosphere expands.... (repeated) cisco J. Varela and Paul Bourgine, editors, Toward a Practice of Autonomous Systems (Proc. of In the generated explanation, the reasoner nds a positive feedbacks among the parameters. This means that ECAL'91), 143{152. The MIT Press(1992). there is a possibility that pheromone atmosphere and 2) Alexis Drogoul, Jacques Ferber, Bruno Corbara, gathering ant population grow exponentially. and Dominique Fresneau: A behavioral simulation The fact that the reasoner can nd such the causal model for the study of emergent social structures, relation in the target system is important, because we In Francisco J. Varela and Paul Bourgine, editors, can use our results in order to redene or re-plan each Toward a Practice of Autonomous Systems (Proc. agent behavior, which the next target of our research. of ECAL'91), 161{170. The MIT Press(1992). The system described in this paper, including the 3) B. Hoelldobler and E.O. Wilson: The Ants, The explanation-generator, is implemented in SICStus Prolog (ver.2.1) on Sun SPARCstation20, which consists of Belknap Press(1990). about 700 clauses. 4) T. S. Ray: An approach to the synthesis of life, In x 5 References C.G. Langton et al., editors, Articial Life II, 371{ Conclusion 408. Addison-Wesley(1991). A method for generating causal network of mobile multi-agent systems is discussed. Our method consists of 1) generating macro-behavior reasoning rules by micro-macro translator, and 2) constructing the causal network among system parameters, in which the directions of change of the parameters are determined, by qualitative simulation with qualitative region grammar. The method is applied to the explanation generation of the foraging behavior of ant colonies. Since the causal network represent the underlying mathematical structures of the target system, the reasoner can nd the macro properties of the target system such as positive/negative feedbacks embedded in the target system, which will be used in redening or re-planning each agent behavior. Our method is powerful enough to reason about many kinds of mobile multi-agent systems including ant colony, cargo-control system, trac navigation, and ow control of industry. 5) R. Axelrod: The Evolution of Cooperation, Basic Books Inc.(1984). 6) K.D. Forbus and B. Falkenhainer: Self-explanatory simulations: An integration of qualitative and quantitative knowledge, In Proc. of AAAI-90(1990) 380{ ) S.A. Rajamoney and S.H. Koo: Qualitative reasoning with microscopic theories, In Proc. of AAAI- 90(1990) 401{406. 8) K.D. Forbus: Qualitative process theory, Articial Intelligence, 24(1984) 85{168. 9) B. Falkenhainer and K.D. Forbus: Setting up largescale qualitative models, In Proc. of AAAI-88(1988) 301{306. {(8){

9 0\F0$9$k%^%k%A!&%(!<%8% %s%h!&%7%9%f% %s%h!&%7%9%f% $NF0:n%"%k%4%j%:% $N<+8J:FJT$K8~$1$ The Authors <VC+ Koichi Kurumatani Cooperative Architecture Project Team He is interested in machine learning and self-reorganization algorithms for multi-agent systems. CfB<??M} Mari Nakamura Life Electronics Research Center She is interested in mathematical biology and neuro-dynamics, especially in bifurcation theory and pattern formation theory on nonlinear distributed systems. {(9){

Qualitative Analysis of Causal Graphs. with Equilibrium Type-Transition 3. LERC, Electrotechnical Laboratory. previous works in the literature.

Qualitative Analysis of Causal Graphs. with Equilibrium Type-Transition 3. LERC, Electrotechnical Laboratory. previous works in the literature. Qualitative Analysis of Causal Graphs with Equilibrium Type-Transition 3 Koichi Kurumatani Electrotechnical Laboratory Umezono 1-1-4, Tsukuba, Ibaraki 305, Japan kurumatani@etl.go.jp Mari Nakamura LERC,

More information

Ant Foraging Revisited

Ant Foraging Revisited Ant Foraging Revisited Liviu A. Panait and Sean Luke George Mason University, Fairfax, VA 22030 lpanait@cs.gmu.edu, sean@cs.gmu.edu Abstract Most previous artificial ant foraging algorithms have to date

More information

Complex Systems Theory and Evolution

Complex Systems Theory and Evolution Complex Systems Theory and Evolution Melanie Mitchell and Mark Newman Santa Fe Institute, 1399 Hyde Park Road, Santa Fe, NM 87501 In Encyclopedia of Evolution (M. Pagel, editor), New York: Oxford University

More information

Deterministic Nonlinear Modeling of Ant Algorithm with Logistic Multi-Agent System

Deterministic Nonlinear Modeling of Ant Algorithm with Logistic Multi-Agent System Deterministic Nonlinear Modeling of Ant Algorithm with Logistic Multi-Agent System Rodolphe Charrier, Christine Bourjot, François Charpillet To cite this version: Rodolphe Charrier, Christine Bourjot,

More information

Mathematical Biology - Lecture 1 - general formulation

Mathematical Biology - Lecture 1 - general formulation Mathematical Biology - Lecture 1 - general formulation course description Learning Outcomes This course is aimed to be accessible both to masters students of biology who have a good understanding of the

More information

An ant colony algorithm for multiple sequence alignment in bioinformatics

An ant colony algorithm for multiple sequence alignment in bioinformatics An ant colony algorithm for multiple sequence alignment in bioinformatics Jonathan Moss and Colin G. Johnson Computing Laboratory University of Kent at Canterbury Canterbury, Kent, CT2 7NF, England. C.G.Johnson@ukc.ac.uk

More information

depending only on local relations. All cells use the same updating rule, Time advances in discrete steps. All cells update synchronously.

depending only on local relations. All cells use the same updating rule, Time advances in discrete steps. All cells update synchronously. Swarm Intelligence Systems Cellular Automata Christian Jacob jacob@cpsc.ucalgary.ca Global Effects from Local Rules Department of Computer Science University of Calgary Cellular Automata One-dimensional

More information

Capacitor Placement for Economical Electrical Systems using Ant Colony Search Algorithm

Capacitor Placement for Economical Electrical Systems using Ant Colony Search Algorithm Capacitor Placement for Economical Electrical Systems using Ant Colony Search Algorithm Bharat Solanki Abstract The optimal capacitor placement problem involves determination of the location, number, type

More information

Engineering Self-Organization and Emergence: issues and directions

Engineering Self-Organization and Emergence: issues and directions 5/0/ Engineering Self-Organization and Emergence: issues and directions Franco Zambonelli franco.zambonelli@unimore.it Agents and Pervasive Computing Group Università di Modena e Reggio Emilia SOAS 005

More information

Introduction to Swarm Robotics

Introduction to Swarm Robotics COMP 4766 Introduction to Autonomous Robotics Introduction to Swarm Robotics By Andrew Vardy April 1, 2014 Outline 1 Initial Definitions 2 Examples of SI in Biology 3 Self-Organization 4 Stigmergy 5 Swarm

More information

Outline. 1 Initial Definitions. 2 Examples of SI in Biology. 3 Self-Organization. 4 Stigmergy. 5 Swarm Robotics

Outline. 1 Initial Definitions. 2 Examples of SI in Biology. 3 Self-Organization. 4 Stigmergy. 5 Swarm Robotics Outline COMP 4766 Introduction to Autonomous Robotics 1 Initial Definitions Introduction to Swarm Robotics 2 Examples of SI in Biology 3 Self-Organization By Andrew Vardy 4 Stigmergy April 1, 2014 5 Swarm

More information

Swarm Intelligence Systems

Swarm Intelligence Systems Swarm Intelligence Systems Christian Jacob jacob@cpsc.ucalgary.ca Department of Computer Science University of Calgary Cellular Automata Global Effects from Local Rules Cellular Automata The CA space is

More information

and combine the results of the searches. We consider parallel search with subdivision, although most notions can be generalized to using dierent searc

and combine the results of the searches. We consider parallel search with subdivision, although most notions can be generalized to using dierent searc On the representation of parallel search in theorem proving Maria Paola Bonacina Department of Computer Science { The University of Iowa Abstract This extended abstract summarizes two contributions from

More information

Math 1270 Honors ODE I Fall, 2008 Class notes # 14. x 0 = F (x; y) y 0 = G (x; y) u 0 = au + bv = cu + dv

Math 1270 Honors ODE I Fall, 2008 Class notes # 14. x 0 = F (x; y) y 0 = G (x; y) u 0 = au + bv = cu + dv Math 1270 Honors ODE I Fall, 2008 Class notes # 1 We have learned how to study nonlinear systems x 0 = F (x; y) y 0 = G (x; y) (1) by linearizing around equilibrium points. If (x 0 ; y 0 ) is an equilibrium

More information

Intuitionistic Fuzzy Estimation of the Ant Methodology

Intuitionistic Fuzzy Estimation of the Ant Methodology BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 9, No 2 Sofia 2009 Intuitionistic Fuzzy Estimation of the Ant Methodology S Fidanova, P Marinov Institute of Parallel Processing,

More information

Using Global Properties for Qualitative Reasoning: A Qualitative System Theory

Using Global Properties for Qualitative Reasoning: A Qualitative System Theory Using Global Properties for Qualitative Reasoning: A Qualitative System Theory Yoshiteru Ishida Division of Applied Systems Science Kyoto University, Kyoto 606 Japan 1 Abstract All the qualitative simulation

More information

Simple Model of Evolution. Reed College FAX (505) Abstract. of evolution. The model consists of a two-dimensional world with a

Simple Model of Evolution. Reed College FAX (505) Abstract. of evolution. The model consists of a two-dimensional world with a Adaptation of Mutation Rates in a Simple Model of Evolution Mark A. Bedau Reed College Robert Seymour 3213 SE Woodstock Blvd., Portland, OR 97202, USA FAX (505) 777-7769 mab@reed.edu rseymour@reed.edu

More information

Comparison of Evolutionary Methods for. Smoother Evolution. 2-4 Hikaridai, Seika-cho. Soraku-gun, Kyoto , JAPAN

Comparison of Evolutionary Methods for. Smoother Evolution. 2-4 Hikaridai, Seika-cho. Soraku-gun, Kyoto , JAPAN Comparison of Evolutionary Methods for Smoother Evolution Tomofumi Hikage, Hitoshi Hemmi, and Katsunori Shimohara NTT Communication Science Laboratories 2-4 Hikaridai, Seika-cho Soraku-gun, Kyoto 619-0237,

More information

Game Theory and Social Psychology

Game Theory and Social Psychology Game Theory and Social Psychology cf. Osborne, ch 4.8 Kitty Genovese: attacked in NY in front of 38 witnesses no one intervened or called the police Why not? \Indierence to one's neighbour and his troubles

More information

biologically-inspired computing lecture 5 Informatics luis rocha 2015 biologically Inspired computing INDIANA UNIVERSITY

biologically-inspired computing lecture 5 Informatics luis rocha 2015 biologically Inspired computing INDIANA UNIVERSITY lecture 5 -inspired Sections I485/H400 course outlook Assignments: 35% Students will complete 4/5 assignments based on algorithms presented in class Lab meets in I1 (West) 109 on Lab Wednesdays Lab 0 :

More information

Book Review John R. Koza

Book Review John R. Koza Book Review John R. Koza Computer Science Department 258 Gates Building Stanford University Stanford, California 94305-9020 USA E-MAIL: Koza@CS.Stanford.Edu WWW: http://www-cs-faculty.stanford.edu/~koza/

More information

Patterns: Observed patterns in nature guide organization and classification and prompt questions about relationships and causes underlying them.

Patterns: Observed patterns in nature guide organization and classification and prompt questions about relationships and causes underlying them. ASET Crosscutting Concepts (CCC) Grade Band Descriptors CCC 1 Patterns: Observed patterns in nature guide organization and classification and prompt questions about relationships and causes underlying

More information

Lesson 5: The Graph of the Equation y = f(x)

Lesson 5: The Graph of the Equation y = f(x) Lesson 5: The Graph of the Equation y = f(x) Learning targets: I can identify when a function is increasing, decreasing, positive and negative and use interval notation to describe intervals where the

More information

hapter 8 Simulation/Realization 8 Introduction Given an nth-order state-space description of the form x_ (t) = f (x(t) u(t) t) (state evolution equati

hapter 8 Simulation/Realization 8 Introduction Given an nth-order state-space description of the form x_ (t) = f (x(t) u(t) t) (state evolution equati Lectures on Dynamic Systems and ontrol Mohammed Dahleh Munther Dahleh George Verghese Department of Electrical Engineering and omputer Science Massachuasetts Institute of Technology c hapter 8 Simulation/Realization

More information

adap-org/ Apr 95

adap-org/ Apr 95 Elements of a Theory of Simulation Steen Rasmussen 1;2 and Christopher L. Barrett 1;2 1 Los Alamos National Laboratory, Los Alamos NM 87545, USA 2 Santa Fe Institute, 1399 Hyde Park Road, Santa Fe NM 87501,

More information

IVR: Introduction to Control (IV)

IVR: Introduction to Control (IV) IVR: Introduction to Control (IV) 16/11/2010 Proportional error control (P) Example Control law: V B = M k 2 R ds dt + k 1s V B = K ( ) s goal s Convenient, simple, powerful (fast and proportional reaction

More information

Measures of Work in Artificial Life

Measures of Work in Artificial Life Measures of Work in Artificial Life Manoj Gambhir, Stephen Guerin, Daniel Kunkle and Richard Harris RedfishGroup, 624 Agua Fria Street, Santa Fe, NM 87501 {manoj, stephen, daniel, rich}@redfish.com Abstract

More information

Introduction to Bioinformatics

Introduction to Bioinformatics Systems biology Introduction to Bioinformatics Systems biology: modeling biological p Study of whole biological systems p Wholeness : Organization of dynamic interactions Different behaviour of the individual

More information

In: Proc. BENELEARN-98, 8th Belgian-Dutch Conference on Machine Learning, pp 9-46, 998 Linear Quadratic Regulation using Reinforcement Learning Stephan ten Hagen? and Ben Krose Department of Mathematics,

More information

Dynamical Embodiments of Computation in Cognitive Processes James P. Crutcheld Physics Department, University of California, Berkeley, CA a

Dynamical Embodiments of Computation in Cognitive Processes James P. Crutcheld Physics Department, University of California, Berkeley, CA a Dynamical Embodiments of Computation in Cognitive Processes James P. Crutcheld Physics Department, University of California, Berkeley, CA 94720-7300 and Santa Fe Institute, 1399 Hyde Park Road, Santa Fe,

More information

How Swarms Build Cognitive Maps

How Swarms Build Cognitive Maps How Swarms Build Cognitive Maps Dante R. Chialvo Mark M. Millonas SFI WORKING PAPER: 1995-03-033 SFI Working Papers contain accounts of scientific work of the author(s) and do not necessarily represent

More information

Ant Colony Optimization: an introduction. Daniel Chivilikhin

Ant Colony Optimization: an introduction. Daniel Chivilikhin Ant Colony Optimization: an introduction Daniel Chivilikhin 03.04.2013 Outline 1. Biological inspiration of ACO 2. Solving NP-hard combinatorial problems 3. The ACO metaheuristic 4. ACO for the Traveling

More information

A Note on the Parameter of Evaporation in the Ant Colony Optimization Algorithm

A Note on the Parameter of Evaporation in the Ant Colony Optimization Algorithm International Mathematical Forum, Vol. 6, 2011, no. 34, 1655-1659 A Note on the Parameter of Evaporation in the Ant Colony Optimization Algorithm Prasanna Kumar Department of Mathematics Birla Institute

More information

agent agent agent... agent

agent agent agent... agent Towards self-organising action selection Mark Humphrys University of Cambridge, Computer Laboratory http://www.cl.cam.ac.uk/users/mh10006 Abstract Systems with multiple parallel goals (e.g. autonomous

More information

Organization of work via the "common stomach" in social insects

Organization of work via the common stomach in social insects Organization of work via the "common stomach" in social insects István Karsai*, Thomas Schmickl** *Dept. Biological Sciences East Tennessee State University **Dept. Zoology Karl Franzens University Graz

More information

Optimal Shape and Topology of Structure Searched by Ants Foraging Behavior

Optimal Shape and Topology of Structure Searched by Ants Foraging Behavior ISSN 0386-1678 Report of the Research Institute of Industrial Technology, Nihon University Number 83, 2006 Optimal Shape and Topology of Structure Searched by Ants Foraging Behavior Kazuo MITSUI* ( Received

More information

Situation. The XPS project. PSO publication pattern. Problem. Aims. Areas

Situation. The XPS project. PSO publication pattern. Problem. Aims. Areas Situation The XPS project we are looking at a paradigm in its youth, full of potential and fertile with new ideas and new perspectives Researchers in many countries are experimenting with particle swarms

More information

Sensitive Ant Model for Combinatorial Optimization

Sensitive Ant Model for Combinatorial Optimization Sensitive Ant Model for Combinatorial Optimization CAMELIA CHIRA cchira@cs.ubbcluj.ro D. DUMITRESCU ddumitr@cs.ubbcluj.ro CAMELIA-MIHAELA PINTEA cmpintea@cs.ubbcluj.ro Abstract: A combinatorial optimization

More information

SWARMING, or aggregations of organizms in groups, can

SWARMING, or aggregations of organizms in groups, can IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS PART B: CYBERNETICS, VOL. 34, NO. 1, FEBRUARY 2004 539 Stability Analysis of Social Foraging Swarms Veysel Gazi, Member, IEEE, and Kevin M. Passino, Senior

More information

Public Key Exchange by Neural Networks

Public Key Exchange by Neural Networks Public Key Exchange by Neural Networks Zahir Tezcan Computer Engineering, Bilkent University, 06532 Ankara zahir@cs.bilkent.edu.tr Abstract. This work is a survey on the concept of neural cryptography,

More information

Nonlinear Dynamics between Micromotives and Macrobehavior

Nonlinear Dynamics between Micromotives and Macrobehavior Nonlinear Dynamics between Micromotives and Macrobehavior Saori Iwanaga & kira Namatame Dept. of Computer Science, National Defense cademy, Yokosuka, 239-8686, JPN, E-mail: {g38042, nama}@nda.ac.jp Tel:

More information

Chapter 1 Introduction

Chapter 1 Introduction 1 Chapter 1 Introduction Figure 1.1: Westlake Plaza A warm sunny day on a downtown street and plaza, pedestrians pass on the sidewalks, people sit on benches and steps, enjoying a cup of coffee, shoppers

More information

3D HP Protein Folding Problem using Ant Algorithm

3D HP Protein Folding Problem using Ant Algorithm 3D HP Protein Folding Problem using Ant Algorithm Fidanova S. Institute of Parallel Processing BAS 25A Acad. G. Bonchev Str., 1113 Sofia, Bulgaria Phone: +359 2 979 66 42 E-mail: stefka@parallel.bas.bg

More information

Pp. 311{318 in Proceedings of the Sixth International Workshop on Articial Intelligence and Statistics

Pp. 311{318 in Proceedings of the Sixth International Workshop on Articial Intelligence and Statistics Pp. 311{318 in Proceedings of the Sixth International Workshop on Articial Intelligence and Statistics (Ft. Lauderdale, USA, January 1997) Comparing Predictive Inference Methods for Discrete Domains Petri

More information

Crosscutting Concepts

Crosscutting Concepts Crosscutting Concepts Patterns Observed patterns of forms and events guide organization and classification, and they prompt questions about relationships and the factors that influence them. Patterns exist

More information

Available online at ScienceDirect. Procedia Computer Science 20 (2013 ) 90 95

Available online at  ScienceDirect. Procedia Computer Science 20 (2013 ) 90 95 Available online at www.sciencedirect.com ScienceDirect Procedia Computer Science 20 (2013 ) 90 95 Complex Adaptive Systems, Publication 3 Cihan H. Dagli, Editor in Chief Conference Organized by Missouri

More information

Pattern Formation and Functionality in Swarm Models

Pattern Formation and Functionality in Swarm Models Pattern Formation and Functionality in Swarm Models Erik M. Rauch Mark M. Millonas Dante R. Chialvo SFI WORKING PAPER: 1995-08-073 SFI Working Papers contain accounts of scientific work of the author(s)

More information

An Integrated Discovery System. Bernd Nordhausen Pat Langley

An Integrated Discovery System. Bernd Nordhausen Pat Langley An Integrated Discovery System Bernd Nordhausen Pat Langley Irvine Computational Intelligence Project Department of Information & Computer Science University of California, Irvine CA 92717 phone : (714

More information

Surrogate regret bounds for generalized classification performance metrics

Surrogate regret bounds for generalized classification performance metrics Surrogate regret bounds for generalized classification performance metrics Wojciech Kotłowski Krzysztof Dembczyński Poznań University of Technology PL-SIGML, Częstochowa, 14.04.2016 1 / 36 Motivation 2

More information

Improving the Performance of Evolutionary Optimization by. Dynamically Scaling the Evaluation Function.

Improving the Performance of Evolutionary Optimization by. Dynamically Scaling the Evaluation Function. Improving the Performance of Evolutionary Optimization by Dynamically Scaling the Evaluation Function Alex S. Fukunaga and Andrew B. Kahng Computer Science Department University of California, Los Angeles

More information

Abstract. Three Methods and Their Limitations. N-1 Experiments Suffice to Determine the Causal Relations Among N Variables

Abstract. Three Methods and Their Limitations. N-1 Experiments Suffice to Determine the Causal Relations Among N Variables N-1 Experiments Suffice to Determine the Causal Relations Among N Variables Frederick Eberhardt Clark Glymour 1 Richard Scheines Carnegie Mellon University Abstract By combining experimental interventions

More information

biologically-inspired computing lecture 12 Informatics luis rocha 2015 INDIANA UNIVERSITY biologically Inspired computing

biologically-inspired computing lecture 12 Informatics luis rocha 2015 INDIANA UNIVERSITY biologically Inspired computing lecture 12 -inspired Sections I485/H400 course outlook Assignments: 35% Students will complete 4/5 assignments based on algorithms presented in class Lab meets in I1 (West) 109 on Lab Wednesdays Lab 0

More information

2 1. Introduction. Neuronal networks often exhibit a rich variety of oscillatory behavior. The dynamics of even a single cell may be quite complicated

2 1. Introduction. Neuronal networks often exhibit a rich variety of oscillatory behavior. The dynamics of even a single cell may be quite complicated GEOMETRIC ANALYSIS OF POPULATION RHYTHMS IN SYNAPTICALLY COUPLED NEURONAL NETWORKS J. Rubin and D. Terman Dept. of Mathematics; Ohio State University; Columbus, Ohio 43210 Abstract We develop geometric

More information

Information Entropy Theory of Physics

Information Entropy Theory of Physics Information Entropy Theory of Physics Abstract... 1 Considering the chess game as a model of physics... 1 Using this model in physics... 4 Quantum Information Science... 4 Quantum Computing Research...

More information

Probability Map Building of Uncertain Dynamic Environments with Indistinguishable Obstacles

Probability Map Building of Uncertain Dynamic Environments with Indistinguishable Obstacles Probability Map Building of Uncertain Dynamic Environments with Indistinguishable Obstacles Myungsoo Jun and Raffaello D Andrea Sibley School of Mechanical and Aerospace Engineering Cornell University

More information

Hidden Markov Models, I. Examples. Steven R. Dunbar. Toy Models. Standard Mathematical Models. Realistic Hidden Markov Models.

Hidden Markov Models, I. Examples. Steven R. Dunbar. Toy Models. Standard Mathematical Models. Realistic Hidden Markov Models. , I. Toy Markov, I. February 17, 2017 1 / 39 Outline, I. Toy Markov 1 Toy 2 3 Markov 2 / 39 , I. Toy Markov A good stack of examples, as large as possible, is indispensable for a thorough understanding

More information

Traffic Signal Control with Swarm Intelligence

Traffic Signal Control with Swarm Intelligence 009 Fifth International Conference on Natural Computation Traffic Signal Control with Swarm Intelligence David Renfrew, Xiao-Hua Yu Department of Electrical Engineering, California Polytechnic State University

More information

Conceptual Modeling in the Environmental Domain

Conceptual Modeling in the Environmental Domain Appeared in: 15 th IMACS World Congress on Scientific Computation, Modelling and Applied Mathematics, Berlin, 1997 Conceptual Modeling in the Environmental Domain Ulrich Heller, Peter Struss Dept. of Computer

More information

Stigmergy: a fundamental paradigm for digital ecosystems?

Stigmergy: a fundamental paradigm for digital ecosystems? Stigmergy: a fundamental paradigm for digital ecosystems? Francis Heylighen Evolution, Complexity and Cognition group Vrije Universiteit Brussel 1 Digital Ecosystem Complex, self-organizing system Agents:

More information

7 Cryptanalysis. 7.1 Structural Attacks CA642: CRYPTOGRAPHY AND NUMBER THEORY 1

7 Cryptanalysis. 7.1 Structural Attacks CA642: CRYPTOGRAPHY AND NUMBER THEORY 1 CA642: CRYPTOGRAPHY AND NUMBER THEORY 1 7 Cryptanalysis Cryptanalysis Attacks such as exhaustive key-search do not exploit any properties of the encryption algorithm or implementation. Structural attacks

More information

2 One-dimensional models in discrete time

2 One-dimensional models in discrete time 2 One-dimensional models in discrete time So far, we have assumed that demographic events happen continuously over time and can thus be written as rates. For many biological species with overlapping generations

More information

Towards Synthesizing Artificial Neural Networks that Exhibit Cooperative Intelligent Behavior: Some Open Issues in Artificial Life Michael G.

Towards Synthesizing Artificial Neural Networks that Exhibit Cooperative Intelligent Behavior: Some Open Issues in Artificial Life Michael G. Towards Synthesizing Artificial Neural Networks that Exhibit Cooperative Intelligent Behavior: Some Open Issues in Artificial Life Michael G. Dyer Computer Science Department, UCLA Overview Introduction

More information

Essentials of Intermediate Algebra

Essentials of Intermediate Algebra Essentials of Intermediate Algebra BY Tom K. Kim, Ph.D. Peninsula College, WA Randy Anderson, M.S. Peninsula College, WA 9/24/2012 Contents 1 Review 1 2 Rules of Exponents 2 2.1 Multiplying Two Exponentials

More information

ξ, ξ ξ Data number ξ 1

ξ, ξ ξ Data number ξ 1 Polynomial Design of Dynamics-based Information Processing System Masafumi OKADA and Yoshihiko NAKAMURA Univ. of Tokyo, 7-- Hongo Bunkyo-ku, JAPAN Abstract. For the development of the intelligent robot

More information

Simulation of cell-like self-replication phenomenon in a two-dimensional hybrid cellular automata model

Simulation of cell-like self-replication phenomenon in a two-dimensional hybrid cellular automata model Simulation of cell-like self-replication phenomenon in a two-dimensional hybrid cellular automata model Takeshi Ishida Nippon Institute of Technology ishida06@ecoinfo.jp Abstract An understanding of the

More information

Reglerteknik, TNG028. Lecture 1. Anna Lombardi

Reglerteknik, TNG028. Lecture 1. Anna Lombardi Reglerteknik, TNG028 Lecture 1 Anna Lombardi Today lecture We will try to answer the following questions: What is automatic control? Where can we nd automatic control? Why do we need automatic control?

More information

Stability theory is a fundamental topic in mathematics and engineering, that include every

Stability theory is a fundamental topic in mathematics and engineering, that include every Stability Theory Stability theory is a fundamental topic in mathematics and engineering, that include every branches of control theory. For a control system, the least requirement is that the system is

More information

bound on the likelihood through the use of a simpler variational approximating distribution. A lower bound is particularly useful since maximization o

bound on the likelihood through the use of a simpler variational approximating distribution. A lower bound is particularly useful since maximization o Category: Algorithms and Architectures. Address correspondence to rst author. Preferred Presentation: oral. Variational Belief Networks for Approximate Inference Wim Wiegerinck David Barber Stichting Neurale

More information

A Cognitive Map for an Artificial Agent

A Cognitive Map for an Artificial Agent A Cognitive Map for an Artificial Agent Unmesh Kurup Rensselaer Polytechnic Institute Carnegie 013, 110 8 th Street Troy, NY 12180 kurupu@rpi.edu B. Chandrasekaran The Ohio State University 591 Dreese,

More information

Complexity 1: Motivation. Outline. Dusko Pavlovic. Introductions. Algorithms. Introductions. Complexity 1: Motivation. Contact.

Complexity 1: Motivation. Outline. Dusko Pavlovic. Introductions. Algorithms. Introductions. Complexity 1: Motivation. Contact. Outline Complexity Theory Part 1: Defining complexity Complexity of algorithms RHUL Spring 2012 Outline Contact Defining complexity email: dusko.pavlovic@rhul.ac.uk phone: (01784 44 30 81) Complexity of

More information

University of California Department of Mechanical Engineering ECE230A/ME243A Linear Systems Fall 1999 (B. Bamieh ) Lecture 3: Simulation/Realization 1

University of California Department of Mechanical Engineering ECE230A/ME243A Linear Systems Fall 1999 (B. Bamieh ) Lecture 3: Simulation/Realization 1 University of alifornia Department of Mechanical Engineering EE/ME Linear Systems Fall 999 ( amieh ) Lecture : Simulation/Realization Given an nthorder statespace description of the form _x(t) f (x(t)

More information

AI Programming CS S-09 Knowledge Representation

AI Programming CS S-09 Knowledge Representation AI Programming CS662-2013S-09 Knowledge Representation David Galles Department of Computer Science University of San Francisco 09-0: Overview So far, we ve talked about search, which is a means of considering

More information

Evolutionary Games and Computer Simulations

Evolutionary Games and Computer Simulations Evolutionary Games and Computer Simulations Bernardo A. Huberman and Natalie S. Glance Dynamics of Computation Group Xerox Palo Alto Research Center Palo Alto, CA 94304 Abstract The prisoner s dilemma

More information

Reasoning with multiple abstraction models

Reasoning with multiple abstraction models Reasoning with multiple abstraction models Yumi Iwasaki Knowledge Systems Laboratory Stanford University 701 Welch Road, Palo Alto, CA 94304 submitted to : Fourth International Workshop on Qualitative

More information

Biology: The Study of Life

Biology: The Study of Life Biology: The Study of Life Before You Read Use the What I Know column to list three things you know about biology. Then list three questions you have about biology in the What I Want to Find Out column.

More information

Structure Learning of a Behavior Network for Context Dependent Adaptability

Structure Learning of a Behavior Network for Context Dependent Adaptability Georgia State University ScholarWorks @ Georgia State University Computer Science Theses Department of Computer Science 12-7-26 Structure Learning of a Behavior Network for Context Dependent Adaptability

More information

Modeling and control of a multi-agent system using mixed integer linear programming

Modeling and control of a multi-agent system using mixed integer linear programming Modeling and control of a multi-agent system using mixed integer linear programming Matthew G. Earl 1 Raffaello D Andrea Abstract The RoboFlag competition was proposed by the second author as a means to

More information

cells [20]. CAs exhibit three notable features, namely massive parallelism, locality of cellular interactions, and simplicity of basic components (cel

cells [20]. CAs exhibit three notable features, namely massive parallelism, locality of cellular interactions, and simplicity of basic components (cel I. Rechenberg, and H.-P. Schwefel (eds.), pages 950-959, 1996. Copyright Springer-Verlag 1996. Co-evolving Parallel Random Number Generators Moshe Sipper 1 and Marco Tomassini 2 1 Logic Systems Laboratory,

More information

The World According to Wolfram

The World According to Wolfram The World According to Wolfram Basic Summary of NKS- A New Kind of Science is Stephen Wolfram s attempt to revolutionize the theoretical and methodological underpinnings of the universe. Though this endeavor

More information

Models of Termite Nest Construction Daniel Ladley Computer Science 2003/2004

Models of Termite Nest Construction Daniel Ladley Computer Science 2003/2004 Models of Termite Nest Construction Daniel Ladley Computer Science 2003/2004 The candidate confirms that the work submitted is their own and the appropriate credit has been given where reference has been

More information

Statistical Analysis of Backtracking on. Inconsistent CSPs? Irina Rish and Daniel Frost.

Statistical Analysis of Backtracking on. Inconsistent CSPs? Irina Rish and Daniel Frost. Statistical Analysis of Backtracking on Inconsistent CSPs Irina Rish and Daniel Frost Department of Information and Computer Science University of California, Irvine, CA 92697-3425 firinar,frostg@ics.uci.edu

More information

Cellular Automata Models of Pedestrian Dynamics

Cellular Automata Models of Pedestrian Dynamics Cellular Automata Models of Pedestrian Dynamics Andreas Schadschneider Institute for Theoretical Physics University of Cologne Germany www.thp.uni-koeln.de/~as www.thp.uni-koeln.de/ant-traffic Overview

More information

September 16, 2004 The NEURON Book: Chapter 2

September 16, 2004 The NEURON Book: Chapter 2 Chapter 2 The ing perspective This and the following chapter deal with concepts that are not NEURON-specific but instead pertain equally well to any tools used for neural ing. Why? In order to achieve

More information

Evolutionary Computation. DEIS-Cesena Alma Mater Studiorum Università di Bologna Cesena (Italia)

Evolutionary Computation. DEIS-Cesena Alma Mater Studiorum Università di Bologna Cesena (Italia) Evolutionary Computation DEIS-Cesena Alma Mater Studiorum Università di Bologna Cesena (Italia) andrea.roli@unibo.it Evolutionary Computation Inspiring principle: theory of natural selection Species face

More information

R eports. Confirmatory path analysis in a generalized multilevel context BILL SHIPLEY 1

R eports. Confirmatory path analysis in a generalized multilevel context BILL SHIPLEY 1 Ecology, 90(2), 2009, pp. 363 368 Ó 2009 by the Ecological Society of America Confirmatory path analysis in a generalized multilevel context BILL SHIPLEY 1 De partement de Biologie, Universite de Sherbrooke,

More information

Changes in properties and states of matter provide evidence of the atomic theory of matter

Changes in properties and states of matter provide evidence of the atomic theory of matter Science 7: Matter and Energy (1) Changes in properties and states of matter provide evidence of the atomic theory of matter Objects, and the materials they are made of, have properties that can be used

More information

Outline. Ant Colony Optimization. Outline. Swarm Intelligence DM812 METAHEURISTICS. 1. Ant Colony Optimization Context Inspiration from Nature

Outline. Ant Colony Optimization. Outline. Swarm Intelligence DM812 METAHEURISTICS. 1. Ant Colony Optimization Context Inspiration from Nature DM812 METAHEURISTICS Outline Lecture 8 http://www.aco-metaheuristic.org/ 1. 2. 3. Marco Chiarandini Department of Mathematics and Computer Science University of Southern Denmark, Odense, Denmark

More information

An explicit dynamic model of segregation

An explicit dynamic model of segregation An explicit dynamic model of segregation Gian-Italo Bischi Dipartimento di Economia e Metodi Quantitativi Universita di Urbino "Carlo Bo" e-mail:gian.bischi@uniurb.it Ugo Merlone Dip. di Statistica e Matematica

More information

A Functional Theory of Design Patterns

A Functional Theory of Design Patterns A Functional Theory of Design Patterns Sambasiva R. Bhatta NYNEX Science & Technology 500 Westchester Ave. White Plains, NY 10604, USA. Ashok K. Goel College of Computing Georgia Institute of Technology

More information

Bayesian Networks Inference with Probabilistic Graphical Models

Bayesian Networks Inference with Probabilistic Graphical Models 4190.408 2016-Spring Bayesian Networks Inference with Probabilistic Graphical Models Byoung-Tak Zhang intelligence Lab Seoul National University 4190.408 Artificial (2016-Spring) 1 Machine Learning? Learning

More information

Solving of logic functions systems using genetic algorithm

Solving of logic functions systems using genetic algorithm Solving of logic functions systems using genetic algorithm V G Kurbanov,2 and M V Burakov Chair of control system of Saint-Petersburg State University of Aerospace Instrumentation, Bolshaya Morskaya, 67,

More information

Knowledge base (KB) = set of sentences in a formal language Declarative approach to building an agent (or other system):

Knowledge base (KB) = set of sentences in a formal language Declarative approach to building an agent (or other system): Logic Knowledge-based agents Inference engine Knowledge base Domain-independent algorithms Domain-specific content Knowledge base (KB) = set of sentences in a formal language Declarative approach to building

More information

Synthesis weakness of standard approach. Rational Synthesis

Synthesis weakness of standard approach. Rational Synthesis 1 Synthesis weakness of standard approach Rational Synthesis 3 Overview Introduction to formal verification Reactive systems Verification Synthesis Introduction to Formal Verification of Reactive Systems

More information

Math Camp Notes: Everything Else

Math Camp Notes: Everything Else Math Camp Notes: Everything Else Systems of Dierential Equations Consider the general two-equation system of dierential equations: Steady States ẋ = f(x, y ẏ = g(x, y Just as before, we can nd the steady

More information

Motivation. Evolution has rediscovered several times multicellularity as a way to build complex living systems

Motivation. Evolution has rediscovered several times multicellularity as a way to build complex living systems Cellular Systems 1 Motivation Evolution has rediscovered several times multicellularity as a way to build complex living systems Multicellular systems are composed by many copies of a unique fundamental

More information

Toward a Better Understanding of Complexity

Toward a Better Understanding of Complexity Toward a Better Understanding of Complexity Definitions of Complexity, Cellular Automata as Models of Complexity, Random Boolean Networks Christian Jacob jacob@cpsc.ucalgary.ca Department of Computer Science

More information

Generating Random Variates II and Examples

Generating Random Variates II and Examples Generating Random Variates II and Examples Holger Füßler Holger Füßler Universität Mannheim Summer 2004 Side note: TexPoint» TexPoint is a Powerpoint add-in that enables the easy use of Latex symbols and

More information

Lectures on Medical Biophysics Department of Biophysics, Medical Faculty, Masaryk University in Brno. Biocybernetics

Lectures on Medical Biophysics Department of Biophysics, Medical Faculty, Masaryk University in Brno. Biocybernetics Lectures on Medical Biophysics Department of Biophysics, Medical Faculty, Masaryk University in Brno Norbert Wiener 26.11.1894-18.03.1964 Biocybernetics Lecture outline Cybernetics Cybernetic systems Feedback

More information

Bacteria Inspired Patterns Grown with Hyperbolic Cellular Automata

Bacteria Inspired Patterns Grown with Hyperbolic Cellular Automata Bacteria Inspired Patterns Grown with Hyperbolic Cellular Automata Maurice Margenstern Université Paul Verlaine Metz, EA3097, and CNRS, LORIA, margens@univ-metz.fr ABSTRACT In this paper we give three

More information

Proceedings of the Seventh Oklahoma Conference on Articial Intelligence, pp , November, Objective Function

Proceedings of the Seventh Oklahoma Conference on Articial Intelligence, pp , November, Objective Function Proceedings of the Seventh Oklahoma Conference on Articial Intelligence, pp. 200-206, November, 1993. The Performance of a Genetic Algorithm on a Chaotic Objective Function Arthur L. Corcoran corcoran@penguin.mcs.utulsa.edu

More information