text stringlengths 128 2.05k |
|---|
[EQUATION] where [MATH] is the [MATH] vector of decision variables within the swarm, which corresponds to the [MATH] SA agent. In the case of multidimensional knapsack problem, the fitness/cost function, [MATH] , corresponds to the objective function (Equation ( )). An agent embedded with fast-track SA explores for bet... |
[EQUATION] where [MATH] is the index for agents, [MATH] and [MATH] represent ” hot ” and ” frozen ” keywords and [MATH] is the problem solving process of the [MATH] agent. There, the improvement towards the optimum value is measured as [MATH] to [MATH] . As expected, the overall search by the whole swarm of SA agents i... |
The whole procedure of coordination by PSO lasts between [MATH] and [MATH] , where [MATH] is the final generation through the whole process. |
Experimental Study This experimental study is not especially to solve multidimensional knapsack problem (MKP), but to test the performance of various approaches including swarm intelligence to coordinate metaheuristic agents. The abovementioned swarm intelligence model for SA agents has been examined with solving multi... |
SA procedure to be run by each agent was investigated for whether to be a 100 iteration long SA to run through 300 generations or a 200 iteration long SA to run 300 generations. The preliminary results confirmed that a 200 iteration long SA algorithm with varying number of generations (Aydin 2008). That was inline with... |
[EQUATION] where [MATH] and [MATH] are the optimum and the average values of experimented results. The average value, [MATH] , is the mean calculated over 50 replications. The second performance measure is the averaged CPU time, which is the mean of the 50 replications. The performance with respect to the solution qual... |
The implementation of the systems has been done using POP C++, which is a GRID programming language developed by Nguyan and Kuonen (2007). It is such a unique distributed programming language that uses object distribution over the targeted infrastructure, and arrange automatic communications among the distributed entit... |
Table presents experimental results with the most fast-track SA agents coordinated with all three approaches against various swarm sizes. The SA algorithm is configured to run 200 iterations without any inner replications, which means that the cooling schedule allows operating once per level of temperature. All three a... |
Table presents the results of experimentations sets which considered 5 inner iterations per SA cycle. These results are much better ones comparing to the single inner iteration case. All three algorithms that coordinate fast-track SA agents, with 5 inner iterations per cycle this time, and improve their performance gra... |
Table shows the experimental results of more focused SA agents, which are replicating 10 times per step of cooling schedule. Since this way of search is more focused, the results of both ESA and PSO hit the optimum 100% with swarm size of 20. Therefore, the experimentation has not proceeded further. As the table manife... |
Fig. indicates the averaged-RPE results of each coordinating approach per benchmark per level of inner iterations in fast-track SA agents. The averaged results are tabulated across horizontal axis pointing out the overall achievement of each approach, where the benchmark problems are indicated as MPK6 and MPK7 with eac... |
Conclusions Metaheuristic agent swarms need collaboration in one way or another to deliver an efficient problem solving services. In this paper, three collaboration algorithms have been examined with respect to efficiency in solution quality. The agents form up the swarms, which are configured as simulated annealing ag... |
Acknowledgements. A part of this study has been carried out in Engineering College of Fribourg in Applied University of Western Switzerland, Fribourg, Switzerland, while the author was visiting GRID research group there. The author is particularly grateful to Prof Pierre Kuonen, the head of GRID research group and Mr. ... |
# Source: arxiv 1308.3400 # Title: Guiding Designs of Self-Organizing Swarms: Interactive and Automated Approaches # Sections: all # Downloaded: 2026-03-03T01:58:19.520905+00:00 |
Guiding Designs of Self-Organizing Swarms: Interactive and Automated Approaches Abstract Self-organization of heterogeneous particle swarms is rich in its dynamics but hard to design in a traditional top-down manner, especially when many types of kinetically distinct particles are involved. In this chapter, we discuss ... |
Introduction Engineering design has traditionally been a top-down process in which a designer shapes, arranges and combines various components in a specific, precise, hierarchical manner, to create an artifact that will behave deterministically in an intended way |
Minai et al., ( 2006 ); Pahl et al., ( 2007 . However, this process does not apply to complex systems that show self-organization, adaptation and emergence. Complex systems consist of a massive amount of simpler components that are coupled locally and loosely, whose behaviors at macroscopic scales emerge partially stoc... |
In an attempt to design engineered complex systems, one of the most challenging problems has been how to bridge the gap between macro and micro scales. Some mathematical techniques make it possible to analytically show such macro-micro relationships in complex systems (e.g., those developed in statistical mechanics and... |
Bar-Yam, ( 2003 ); Doursat et al., ( 2012 . Unfortunately, such cases are exceptions in a vast, diverse, and rather messy compendium of complex systems dynamics Camazine, ( 2003 ); Sole and Goodwin, ( 2008 . To date, the only generalizable methodology available for predicting macroscopic properties of a complex system ... |
More importantly, the other way of connecting the two scales—embedding macroscopic requirements the designer wants into microscopic rules that will collectively achieve those requirements—is by far more difficult. This is because the mapping between micro and macro scales is highly nonlinear, and also the space of poss... |
Dawkins, ( 1996 for complex systems design is empirically supported by the fact that it has been the primary mechanism that has produced astonishingly complex, sophisticated, highly emergent machinery in the history of real biological systems. |
The combination of these two methodologies—experiment and evolution—that connect macro and micro scales in two opposite directions (the whole cycle in Fig. ) is now a widely adopted approach for guiding systematic design of self-organizing complex systems Minai et al., ( 2006 ); Anderson, ( 2006 . Typical design steps ... |
Such experiment-and-evolution-based design of complex systems is not free from limitations, however. In typical evolutionary design methods, the designer needs to explicitly define a performance metric, or “fitness”, of design candidates, i.e., how good a particular design is. Such performance metrics are usually based... |
In this chapter, we present our efforts to address this problem, by (1) utilizing and enhancing interactive evolutionary design methods and (2) realizing spontaneous evolution of self-organizing swarms within an artificial ecosystem. |
Model: Swarm Chemistry We use Swarm Chemistry Sayama, ( 2007 2009 as an example of self-organizing complex systems with which we demonstrate our design approaches. Swarm Chemistry is an artificial chemistry |
Dittrich et al., ( 2001 model for designing spatio-temporal patterns of kinetically interacting heterogeneous particle swarms using evolutionary methods. A swarm population in Swarm Chemistry consists of a number of simple particles that are assumed to be able to move to any direction at any time in a two- or three-dim... |
If there are no other particles within its local perception range, steer randomly ( Straying ). Otherwise: Steer to move toward the average position of nearby particles Cohesion , Fig. (a)). |
Steer toward the average velocity of nearby particles ( Alignment , Fig. (b)). Steer to avoid collision with nearby particles ( Separation , Fig. (c)). |
Steer randomly with a given probability ( Randomness ). Approximate its speed to its own normal speed ( Self-propulsion ). These rules are implemented in a simulation algorithm that uses kinetic parameters listed and explained in Table (see |
Sayama, ( 2009 2010 for details of the algorithm). The kinetic interactions in our model uses only one omni-directional perception range ( [MATH] ), which is much simpler than other typical swarm models that use multiple and/or directional perception ranges |
Reynolds, ( 1987 ); Couzin et al., ( 2002 ); Kunz and Hemelrijk, ( 2003 ); Hemelrijk and Kunz, ( 2005 ); Cheng et al., ( 2005 ); Newman and Sayama, ( 2008 . Moreover, the information being shared by nearby particles is nothing more than kinetic one (i.e., relative position and velocity), which is externally observable ... |
Each particle is assigned with its own kinetic parameter settings that specify preferred speed, local perception range, and strength of each kinetic rule. Particles that share the same set of kinetic parameter settings are considered of the same type. Particles do not have a capability to distinguish one type from anot... |
For a given swarm, specifications for its macroscopic properties are indirectly and implicitly woven into a list of different kinetic parameter settings for each swarm component, called a recipe |
(Fig. Sayama, ( 2009 . It is quite difficult to manually design a specific recipe that produces a desired structure and/or behavior using conventional top-down design methods, because the self-organization of a swarm is driven by complex interactions among a number of kinetic parameters that are intertwined with each o... |
In the following sections, we address this difficult design problem using evolutionary methods. Unlike in other typical evolutionary search or optimization tasks, however, in our swarm design problem, there is no explicit function or algorithm readily available for assessing the quality (or fitness) of each individual ... |
Interactive Approach The first approach is based on interactive evolutionary computation (IEC) Banzhaf, ( 2000 ); Takagi, ( 2001 , a derivative class of evolutionary computation which incorporates interaction with human users. Most IEC applications fall into a category known as “narrowly defined IEC” (NIEC) |
Takagi, ( 2001 , which simply outsources the task of fitness evaluation to human users. For example, a user may be presented with a visual representation of the current generation of solutions and then prompted to provide fitness information about some or all of the solutions. The computer in turn uses this fitness inf... |
Our initial work, Swarm Chemistry 1.1 Sayama, ( 2007 2009 , also used a variation of NIEC, called Simulated Breeding Unemi, ( 2003 . This NIEC-based application used discrete, non-overlapping generation changes. The user selects one or two favorable swarms out of a fixed number of swarms displayed, and the next generat... |
As a design tool, NIEC has some disadvantages. One set of disadvantage stems from the confinement of the user to the role of selection operator (Fig. , left). Creative users who are accustomed to a more highly involved design process may find the experience to be tedious, artificial, and frustrating. Earlier literature... |
Bentley and O’Reilly, ( 2001 and that the users should be the initiators of actions rather than simply responding to prompts from the system |
Shneiderman et al., ( 2009 These lines of research suggest that enhancing the level of interaction and control of IEC may help the user better guide the design process of self-organizing swarms. Therefore, we developed the concept of hyperinteractive evolutionary computation (HIEC) |
Bush and Sayama, ( 2011 , a novel form of IEC in which a human user actively chooses when and how to apply each of the available evolutionary operators, playing the central role in the control flow of evolutionary search processes (Fig. , right). In HIEC, the user directs the overall search process and initiates action... |
We developed Swarm Chemistry 1.2 Sayama et al., ( 2009 ); Bush and Sayama, ( 2011 , a redesigned HIEC-based application for designing swarms. This version uses continuous generation changes, i.e., each evolutionary operator is applied only to part of the population of swarms on a screen without causing discrete generat... |
Sayama et al., ( 2009 ); Bush and Sayama, ( 2011 We conducted the following two human-subject experiments to see if HIEC would produce a more controllable and positive user experience, and thereby better swarm design outcomes, than those with NIEC. |
3.1 User experience In the first experiment, individual subjects used the NIEC and HIEC applications mentioned above to evolve aesthetically pleasing self-organizing swarms. We quantified user experience outcomes using questionnaire, in order to quantify potential differences in user experience between the two applicat... |
Twenty-one subjects were recruited from students and faculty/staff members at Binghamton University. Each subject was recruited and participated individually. The subject was told to spend five minutes using each of two applications to design an “interesting and lifelike” swarm. Each of these two applications ran on th... |
The results are shown in Fig. . Of the 7 factors measured, 3 showed statistically significant difference between two platforms: controllability, fun factor, and overall satisfaction. The higher controllability ratings for HIEC suggest that our original intention to re-design an IEC framework to grant greater control to... |
3.2 Design quality The goal of the second experiment was to quantify the difference between HIEC and NIEC in terms of final design quality. In addition, the effects of mixing and mutation operators on the final design quality were also studied. The key feature of this experiment was that design quality was rated not in... |
Twenty-one students were recruited for this experiment. Those subjects did not have any overlap with the subjects of experiment 1. The subjects were randomly divided into groups of three and instructed to work together as a team to design an “interesting” swarm design in ten minutes using either the NIEC or HIEC applic... |
The result is shown in Fig. . There was a difference in the average rating scores between designs created using NIEC and HIEC (conditions 0 and 4), and the rating scores were higher when more evolutionary operators were made available. Several final designs produced through the experiment are shown in Fig. |
(three with the highest scores and three with the lowest scores), which indicate that highly evaluated swarms tended to maintain coherent, clear structures and motions without dispersal, while those that received lower ratings tended to disperse so that their behaviors are not appealing to students. |
To detect statistical differences between experimental conditions, a one-way ANOVA was conducted. The result of the ANOVA is summarized in Table . Statistically significant variation was found between the conditions ( [MATH] ). Tukey’s and Bonferroni’s post-hoc tests detected a significant difference between conditions... |
Automated Approach The second approach we took was motivated by the following question: Do we really need human users in order to guide designs of self-organizing swarms? This question might sound almost paradoxical, because designing an artifact implies the existence of a designer by definition. However, this argument... |
Dawkins, ( 1996 . Now that we know that the blind evolutionary process did “design” quite complex, intricate structures and functions of biological systems, it is reasonable to assume that it should be possible to create automatic processes that can spontaneously produce various creative self-organizing swarms without ... |
In order to make the swarms capable of spontaneous evolution within a simulated world, we implemented several major modifications to Swarm Chemistry Sayama, ( 2010 2011 ); Sayama and Wong, ( 2011 , as follows: |
1. There are now two categories of particles, active (moving and interacting kinetically) and passive (remaining still and inactive). An active particle holds a recipe of the swarm (a list of kinetic parameter sets) (Fig. (a)). |
2. A recipe is transmitted from an active particle to a passive particle when they collide, making the latter active (Fig. (b)). |
3. The activated particle differentiates randomly into one of the multiple types specified in the recipe, with probabilities proportional to their ratio in it (Fig. (c)). |
4. Active particles randomly and independently re-differentiate with small probability, [MATH] , at every time step ( [MATH] for all simulations presented in this chapter). |
5. A recipe is transmitted even between two active particles of different types when they collide. The direction of recipe transmission is determined by a competition function that picks one of the two colliding particles as a source (and the other as a target) of transmission based on their properties (Fig. (d)). |
6. The recipe can mutate when transmitted, as well as spontaneously at every time step, with small probabilities, [MATH] and [MATH] respectively (Fig. (e)). In a single recipe mutation event, several mutation operators are applied, including duplication of a kinetic parameter set (5% per set), deletion of a kinetic par... |
These extensions made the model capable of showing morphogenesis and self-repair Sayama, ( 2010 and autonomous ecological/evolutionary behaviors of self-organized “super-organisms” made of a number of swarming particles Sayama, ( 2011 ); Sayama and Wong, ( 2011 . We note here that there was a technical problem in the o... |
Sayama, ( 2011 , which was fixed in the later implementation Sayama and Wong, ( 2011 In addition, in order to make evolution occur, we needed to confine the particles in a finite environment in which different recipes compete against each other. We thus conducted all the simulations with 10,000 particles contained in a... |
In the simulations, two different initial conditions were used: a random initial condition made of 9,900 inactive particles and 100 active particles with randomly generated one-type recipes distributed over the space, and a designed initial condition consisted of 9,999 inactive particles distributed over the space, wit... |
4.1 Exploring experimental conditions Using the evolutionary Swarm Chemistry model described above, we studied what kind of experimental conditions (competition functions and mutation rates) would be most successful in creating self-organizing complex patterns Sayama, ( 2011 |
The first experiment was to observe the basic evolutionary dynamics of the model under low mutation rates ( [MATH] [MATH] ). Random and designed (“swinger”) initial conditions were used. The following four basic competition functions were implemented and tested: |
faster : The faster particle wins. slower : The slower particle wins. behind : The particle that hit the other one from behind wins. Specifically, if a particle exists within a 90-degree angle opposite to the other particle’s velocity, the former particle is considered a winner. |
majority : The particle surrounded by more of the same type wins. The local neighborhood radius used to count the number of particles of the same type was 30. The absolute counts were used for comparison. |
Results are shown in Fig. 10 . The results with the “behind” competition function were very similar to those with the “faster” competition function, and therefore omitted from the figure. In general, growth and replication of macroscopic structures were observed at early stages of the simulations. The growth was accomp... |
It was also observed that the choice of competition functions had significant impacts on the system’s evolutionary dynamics. Both the “faster” and “behind” competition functions always resulted in an evolutionary convergence to a homogeneous cloud of fast-moving, nearly independent particles. In contrast, the “slower” ... |
Based on the results of the previous experiment, the following five more competition functions were implemented and tested. The last three functions that took recipe length into account were implemented in the hope that they might promote evolution of increasingly more complex recipes and therefore more complex pattern... |
majority (probabilistic) : The particle surrounded by more of the same type wins. This is essentially the same function as the original “majority”, except that the winner is determined probabilistically using the particle counts as relative probabilities of winning. |
majority (relative) : The particle that perceives the higher density of the same type within its own perception range wins. The density was calculated by dividing the number of particles of the same type by the total number of particles of any kind, both counted within the perception range. The range may be different a... |
recipe length : The particle with a recipe that has more kinetic parameter sets wins. recipe length then majority : The particle with a recipe that has more kinetic parameter sets wins. If the recipe length is equal between the two colliding particles, the winner is selected based on the “majority” competition function... |
recipe length [MATH] majority : A numerical score is calculated for each particle by multiplying its recipe length by the number of particles of the same type within its local neighborhood (radius = 30). Then the particle with a greater score wins. |
Results are summarized in Fig. 11 . As clearly seen in the figure, the majority-based rules are generally good at maintaining macroscopic coherent structures, regardless of minor variations in their implementations. This indicates that interaction between particles, or “cooperation” among particles of the same type to ... |
In the meantime, the “recipe length” and “recipe length then majority” competition functions did not show any evolution toward more complex forms, despite the fact that they would strongly promote evolution of longer recipes. What was occurring in these conditions was an evolutionary accumulation of “garbage” kinetic p... |
The results described above suggested the potential of evolutionary Swarm Chemistry for producing more creative, continuous evolutionary processes, but none of the competition functions showed notable long-term evolutionary changes yet. We therefore increased the mutation rates to a 100 times greater level than those i... |
Salzberg et al., ( 2004 ); Salzberg and Sayama, ( 2004 , which demonstrated that such dynamic environments may make evolutionary dynamics of a system more variation-driven and thus promote long-term evolutionary changes. |
With these additional changes, some simulation runs finally demonstrated continuous changes of dominant macroscopic structures over a long period of time (Fig. 12 ). A fundamental difference between this and earlier experiments was that the perturbation introduced to the environment would often break the “status quo” e... |
4.2 Quantifying observed evolutionary dynamics The experimental results described above were quite promising, but they were evaluated only by visual inspection with no objective measurements involved. To address the lack of quantitative measurements, we developed and tested two simple measurements to quantify the degre... |
Evolutionary exploration was quantified by counting the number of new RGB colors that appeared in a bitmap image of the simulation snapshot at a specific time point for the first time during each simulation run (Fig. 13 , right). Since different particle types are visualized with different colors in Swarm Chemistry, th... |
We applied these measurements to simulation runs obtained under each of the four conditions shown in Table . Results are summarized in Figs. 14 and 15 . Figure 14 clearly shows the high evolutionary exploration occurring under the conditions with high mutation rates and environmental perturbations. In the meantime, Fig... |
Sayama, ( 2011 ); Sayama and Wong, ( 2011 that mistakenly depended on perception ranges of particles. The “revised” conditions used a fixed collision detection algorithm. This modification was found to have an effect to maintain macroscopic structures for a prolonged period of time (Fig. 15 ). Combining these results t... |
Conclusions In this chapter, we have reviewed our recent work on two complementary approaches for guiding designs of self-organizing heterogeneous swarms. The common design challenge addressed in both approaches was the lack of explicit criteria for what constitutes a “good” design to produce. In the first approach, th... |
The core message arising from both approaches is the unique power of evolutionary processes for designing self-organizing complex systems. It is uniquely powerful because evolution does not require any macroscopic plan, strategy or global direction for the design to proceed. As long as the designer—this could be either... |
We conclude this chapter with a famous quote by Richard Feynman. At the time of his death, Feynman wrote on a blackboard, “What I cannot create, I do not understand.” This is a concise yet profound sentence that beautifully summarizes the role and importance of constructive understanding (i.e., model building) in scien... |
Acknowledgments We thank the following collaborators and students for their contributions to the research presented in this chapter: Shelley Dionne, Craig Laramee, David Sloan Wilson, J. David Schaffer, Francis Yammarino, Benjamin James Bush, Hadassah Head, Tom Raway, and Chun Wong. This material is based upon work sup... |
# Source: arxiv 1309.1524 # Title: Guided Self-Organization of Input-Driven Recurrent Neural Networks # Sections: all # Downloaded: 2026-03-03T01:58:17.851746+00:00 |
Guided Self-Organization of Input-Driven Recurrent Neural Networks Introduction To understand the world around us, our brains solve a variety of tasks. One of the crucial functions of a brain is to make predictions of what will happen next, or in the near future. This ability helps us to anticipate upcoming events and ... |
A marked feature of brain networks is the massive amount of recurrent connections between cortical areas, especially on a local scale (Douglas et al.,, 2004 . Since information in these recurrent connections, or loops, can circulate between many neurons in a given circuit, they are ideally suited to provide a time-cont... |
In RNNs, many detailed properties of real neurons are abstracted for the sake of tractability, but important general concepts are kept. Elements of these networks are simple nodes that combine inputs from other nodes in the network, usually in a nonlinear fashion, to form their outputs. They are connected in a directed... |
While the recurrent connections of RNNs enable them to deal with time-dependencies in the input data, they also complicate training procedures compared to algorithms for networks without loops (e.g., Backpropagation Rumelhart et al., ( 1986 or R-Prop Riedmiller and Braun, ( 1993 ). Notably, training RNNs with tradition... |
The realization of these fundamental issues led to alternative ways of using and training RNNs, some of which can be summarized in the field of Reservoir Computing methods (see, e.g., a recent overview by Lukoševičius and Jaeger,, 2009 , specialized architectures like the Long Short Term Memory (LSTM) networks Hochreit... |
Reservoir methods implement a fixed high-dimensional reservoir of neurons, using random connection weights between the hidden units, chosen small enough to guarantee dynamic stability. Input weights into this reservoir are also selected randomly, and reservoir learning procedures train only the output weights of the ne... |
Taking the echo state network approach as a specific example of a typical reservoir network (see Fig. ), it will consist of the following components: A random input-matrix [MATH] combines input values [MATH] linearly and sends them to the units in the high-dimensional hidden layer, referred to as the reservoir . The un... |
[EQUATION] The input and hidden layer connections, [MATH] and [MATH] , are not trained in reservoir computing approaches. It is also possible to introduce feedback connections from outputs back into the reservoir Lukoševičius and Jaeger, ( 2009 . To approximate a specific target function, only the output weights [MATH]... |
This approach works very well in practice. However, results will depend on the particular random set of weights that is drawn. In fact, there is considerable variability in performance between runs of networks with equal parameter settings, but different reservoir weights drawn each time Ozturk et al., ( 2007 . Strikin... |
Driving self-organization into a desired direction requires an understanding what properties a good RNN or reservoir network has. The mathematical tools to understand computation in these networks (which are instances of the larger class of input-driven dynamical systems) are still under active development Manjunath et... |
In this chapter, we review attempts that have been made towards understanding the computational properties and mechanisms of input-driven dynamical systems like RNNs, and reservoir computing networks in particular. We provide details on methods that have been developed to give quantitative answers to the questions abov... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.