Ya he instalado un plugin que me permita escribir directamente fórmulas en 
Voy a probarlo un tiempo, a ver qué tal va. De momento parece sencillo y espero que sea rápido renderizando las fórmulas. Creo que me servirá; no pretendo escribir fórmulas especialmente complicadas.
Si no sirve, otra posibilidad es LatexRender, pero esta es más compleja y requiere algunas aplicaciones en el servidor web y tocar algo ficheros de WordPress, lo que no me gusta demasiado.
Well, it’s Friday afternoon and these are the latest sessions of EUMAS. None of them are specially interesting for me.
Session 7 is about simulation.
The two first articles are presented by a third person, not by their authors. And the last author hasn’t appeared yet. It seems that they are simulated authors too :-D
It begins with an article about simulation with virtual lanes for motorcycles simulation, trough microscopic simulation. The problem of current approaches is that they assume that motorcycles rides in the middle of a lane. I’ve never seen a cycle been driven like that ;-) and so do they, so they plan a more realistic scene, in which the cycle can change the lanes and uses the road marking as valid paths. That is: they are driven as motorcycles. In their proposal, each vehicle is an agent and the traffic is just an emergent behaviour of the multiagent system. Usually, only road is used, but in cases of dense traffic, all free space is used. This “new” free spaces for cycles are the virtual lanes (VLs). VLs can be authorised or forbidden depending on the position of the rest of vehicles (due to their proximity to the limits of the actual lane).Several rules to model the behaviour of the cycles are proposed (to pass on the left, to drive in the widest lane and so on).They need a deeper validation to compare with real data in the behaviour of traffic conditions and individual driver’s behaviour, and to simulate complex situations, as junctions.
Now it’s Elena’s turn, with the Gus’s paper Intelligent Agents in Serious Games. After the introduction, it classifies the different approaches in four types:
The problem is that there is not an agent model common for all these approaches so,tachaaaan, we propose our unified agent model for serious game: Spade (see references). A couple of questions, but Elena can’t answer them, so they will be redirected to garanda {at} dsic.upv.es X-D
The third author is virtual already, so we’ve finished.
I’ve arrived late to the first one, but it sounds a bit rare. I haven’t understood their purpose: testing agents for bugs in their codification? I’m not sure.
The second one is about security in mobile agent systems for open environment. Current solutions, as DNS or FIPA-AMS are not usefull. Agent location service requires to be efficient, by dealin with dynamic information and to be scalable, and to be secure. And they are looking for a model that provide transparent location for agents. They’ve consider several alternatives–(de)centralised, hierarchical/P2P– The solution is a Fonkey-based location server. The idea is that every information have to be signed. It is a platform to provide secure communications for agent platforms. May be is what we get with Kerberos for Magentix (some how). At the end, he recognises that send packages in DHT is expensive.
Now is Jose Miguel’s turn with Magentix. Only sockets can be used to communicate processes in different machines, so the efficiency of communication among agents depends on the efficiency of socket management. Comparing Java with TCP-based pure communication, the former is clearly the loser (obviuos). After that, he continues explaining the Magentix agent platform and its performance. Any questions? Yes. The first one is about the scalability of AMS, because white pages are all replicated in all hosts. Other question is that we’re are making platform dependent something that is platform independent. And a pair of questions more. A lot of criticism; I guess that Magentix hasn’t convinced.
The last paper is about a methodology called ASEME. The analyse the systems using capabilities and functionalities., The design phase models the intra-agent and inter-agent control. The design phase ends with a platform independent model (PIM) and uses PSMs in the implementation phase (I have to review again our MDA-based proposal). Till now, it is just another methodology. I think that agent community needs a common methodology (as UML for OOP) and its own high level language. I’m bored of different methodologies (our own methodologies are included). Something interesting is the use of statecharts for representing protocols, because its more important to establish the activities that the agent have to do when a message arrives. Furthermore, it can be easily linked with the rest of agent activities.
Some that I have to read is something about SPEM. It seems to be a notation to describe the different phases of a methodology. Maybe it will be interesting for Thomas.
Well, and here finishes the 5th European Conference on Multiagent Systems. I’ll summarise the conference in one last post.
Greetings from Hammamet.
Blogged with Flock
Good morning. Another day in EUMAS, after a stressing dinner. I’ll talk about that, because now its time to work. The topics for today are not specially interesting for me.
Semantic Alignment of Agent Interactions trough the Communication Protocol explores how to use semantical alignment using communication as a complement to other techniques. They assume that (i) all agents participate in the same kind of interaction, (ii) they’re in the same state on the transaction. Interesting, but a bit immature.
And the last author has disappeared. This is becoming a disgusting costume in the EUMAS. I haven’t seen the semantic web nor the e-commerce anywhere. A little deception.
Now its time for a coffee, see you later.
A hard (a very hard) session for me. This is the kind of papers that I’d like to write when I get old. I haven’t anything to write. Just to see, to listen and to try to understand something. One more thing: I need a LaTeX plugin for wordpress. I’ll add that to the todo list.
Blogged with Flock
A very interesting speech about how autonomous agents can coordinate plans, balancing the plan construction and its coordinate execution. A good introduction to the topic.
1. plan coordination methods
He begins with an easy example in which two agents fall in a deadlock if they have to coordinate their tasks but they have their own preferences about the order in which the agents want its actions to be executed (I’ve seen this problem yesterday in a session). So the problem is that we have a set of autonomous agents, a partially ordered set of tasks, own preferences and dependences among the agents. How can we solve this coordination problem?
2. minimun coordination methods
The before planning problem you have a set T of independent tasks distributed into a set of autonomous agents A where the agents plans automously. We enforce the agents to eliminate those constraints that create cycles in the coordinate plan. But this problem is complex to solve.
First, how we can verify that a plan is coordinated? Two techniques: fixed coordination verif. and free coordination verif. both at least are co-NP complete problems :-(
3. efficient methods for approximating minimum coord
They are hard problems: EFixC is still NP-complete for agents with 2 tasks, but there are heuristics,as label setting algorithm, to reduce the cost.
The label setting alg. adds the tasks in “layers”, beginning with tasks without precedents agent by agent. All arcs (dependences) among tasks always will be from lower layers to higher ones, so we can ensure that we are avoiding inter-agent cycles. It works particularlly well with certain structures. (i) if the agents form a chain with length k, the (ii) if agents form an star, with a central “server”.
4. the price of autonomy
Individual agents pay a price for their autonomy in order to get plans coordinated. as the division between the sum cost of an optimal local plan / the cost of the optimal global, joint plan. If this value is too high it is not worthy.
5. two applications:
multimodal transportation
We have to order a sequence of simple transportation tasks between to locations in two cities. The cost of the plan is measured by #moves + #(un)load actions. They’ve compared the coordination approach over benchmark set with planners that treat the whole problem and with planners that use local planners (for each agent) plus coordination (STAN, TAL and HSP among others). The conclusion is that if they combine HSP with their min. coord. method, the efficiency of the planner is increased dramatically.
context aware routing
We’re running out of time, so he’s not going to explain with detail the second example. They’ve compared fixed path scheduling with time windows path routing. The last one has better results.
To conclude…
The pre-planning coordination is suitable for coordinate autonomous planning agents. This approach decompose complex problems in minimal changing instances. The price of the autonomy determines the quality of the decomposition.
It’s been a good talk. I’ve remembered past times… and it’s given be some goods ideas to be used in the discovering and composition of semantic web services using hypergraphs and HSP-based planning algorithms.
Blogged with Flock
This afternoon I’m going to attend the session about web services and another one about coordination and negotiation.
The room is quite empty, only 10 people. The session begins with our paper Service Discovery an Composition in Multiagent Systems. It’s a review about the most common techniques and algorithms used for standard web services and agent-based solutions. A future works point had to be added, and that has been the first question: ok, a good revision, and now… what are you planning to do? :-) Well, to integrate in a model of service-oriented, agent-based, organisation-focused architecture for open systems. It sounds good but we have to begin working. Current solutions that use gateways, as WSIG or Agent2WS aren’t useful. We need a common language for agents and services to live together in the same platform. Both can provide services (agent services may be more complex) and agents will be responsible of dynamic service composition.
The room is getting crowd, we’re 25 people now and the second talk begins. There’s a problem with his computer, so the chair exchanges 2nd and 3rd talks: Towards a Modular Architecture of Argumentative Agents to Compose Services. A good approach that refers to services in general: not web services, but services provided by agents that form a virtual organisation. The classify the demands according to different dialog types (information-seek, negotiate…). I have to check it with the dialog classification that I used in the AIWS course. They propose three models
(I like it) An interesting idea: to use priorities among goals and decisions. A good paper but, at the end, the negotiation of the QoS parameters is done when they have a list of winners, that is, when they have identified the service. Can be this idea be used during the discovery of the service? That would allow to discard those services that never meet the user QoS parameters.
And, finally the last one: Agent-based Framework for Web Service Composition. Agggg! his using comic font… I’m not sure that could take it seriously ;-) Well, a CBR-based solution to compose web services is proposed, so they can use semantic information to match the services (syntactic distance).I can’t see the relationship with agents. Wait a moment, I have it. It’s because they use a distributed solution.
And that’s the end of the session, with 30 people in the room. An active session, with a lot of questions. If this is the rhythm of the EUMAS, it is going to be an interesting workshop.
After the coffee-break, the last session of the day. The first paper, Plan Coordination for Durative Tasks, is very similar to some things that I’ve written in my PhD. thesis. I have to read it carefully. Furthermore, Cees Witteveen is the coauthor and he has an invited talk tomorrow. I can’t miss it.
The second talk, Bilateral Agent Negotiation With Information-Seeking, defines a framework to argumentative negotiation to an specific resource allocation case. The agents have (i) beliefs, (ii) desires, (iii) actions, that can be external communications or internal–message processing–, (iv) messages, (v) commitments to beliefs, desires and dialogs, (vi) action rules–actions, preconditions, constraints and consequences–, (vii) an evaluation mechanism to determine agent’s intentions and (viii) preferences as priorities over actions. They use protocols for 2 types of dialogs: for information-seeking (query and response) and negotiation( offer, accept and reject). And the permissible messages, the turn taking and the order in the messages are defined for both. After the model is exposed, he shows how they can solve the resource allocation problem with this technique, but there’re some cases without a soloution (cycles or agents that not need anything) because the negotiation process is blocked by self-interested behaviours of the agent.
The third paper is Teamwork Coordination for Vehicle Routing Problem and tries to find the simplest global decision that produces the max. global utility. They solve it using statistical methods from IR (maximum entropy). Sometimes, the utility is obtained not by one action, but by a sequence of actions. Furthermore, some immediate good actions for an agent can be bad for the global final utility, so they have to be punished asap.
Another article about traffic conditions is Anticipatory Vehicle Routing Using Delegate Multiagent Systems, which tries to anticipate and avoid congestions. the road is modelled as a graph where the nodes are the road junctions. Each node has one infrastructure agent and each car a vehicle agent. Before the car arrives a node, the vehicle sends exploratory ants, which ask infrastructure agents for the estimated time for their part of the route. Hundreds of ants can be sent by different vehicles. Intention ants are sent to tentative book a route. The booking have to be refreshed continuously (booking decay). Interesting solution, but with a lot of problems. The most important one is the fairness: I can book every possible route, so the estimate traffic can be faked.
Blogged with Flock
Tags: eumas, congreso, web services, cbr, coordinacion, negociacion
Well, I’ll summarise some interesting talks that I’ve seen this morning. In general, all participants seems to be interested in the presentations: a lot of questions that the chair has to interrupt to maintain the session on time. I like it, more that the one in CAEPIA, in Salamanca, some weeks ago (I talked about that).
I’ve attended two sessions: one about trust and reputation and the other about organisations and institutions. As I twittered, the first speaker in the trust session hasn’t appeared, so I had the opportunity of see an interesting speech about agents with hair. You’ll understand that when you see the pictures :-).
A Common Basis for Agents Organisations in BDI Language.
(Also knows as hairy agents :-) Defines a basic agent as a tuple <S, SP> where S is the state and SP the behavioural specifications. And now they extends the state S with the content (Ct) and the context (Cx) of the agent. With this approach, is interesting to see how two agents intersect or how agents can be embedded into others, allowing us to define organisations.
After this paper, we continue in the trust session with the following two talks.
Supporting Experimentation for Trust Models in Virtual Organisations: TOAST
Blogged with Flock
The EUMAS ’07 begins with the invited talk of Leila Amgoud titles “Modelling Negotiation Dialogs Using Argumentation”.
At the moment, it’s been a very basic talk explaining what’s negotiation between autonomous agents. After explain what negotiation is, she talks about the approaches to the negotiation process: game theory, heuristics and argumentation based approaches.
Game theory
As a branch of economics, assumes perfect rationality (all information is know), not an actual case. It not says anything about how the utility function is calculated and the agents can only emit proposals.
Heuristic-based
Agents don’t know each other preferences, but they only exchange proposals and the preferences of each agent are fixed.
Argumentation-based
We can influence the preference relation of other agents, because we interchange proposals and arguments to “convince” the other agent.
l
Now she continues explaining things about argumentation. Differentiates between two types of arguments: epistemic and practical. The former supports beliefs and is used in deductive reasoning, whereas the latest is based on proposals and is used in abductive reasoning. It’s explaining basic things again (attacks, conflict-free sets…). Arguments can be credulitily or skeptically accepted or rejected. To the argumentation process take into account only the skeptical ones or the rejected to order the offers.
And now…. begins the interesting part: how to use argumentation in negotiation processes (it has taken her 40 min.!!)
She defines 3 protocol classes, but I hardy can read the slides.
… and 3 types of strategies depending on
Well. this where the important part of the conference and she have used just 2 slides :-( Instead of explaining the strategies, she’s continuing with an example. She compares one example without argumentation. The agents can be jammed, but the problem is solve if they can argument about their preferences (obviously).
The problem is if an optimal solution can be reached quikly. But I thing that this is not too importal. Actual life is full of non-optimal negotiations, but good enough ones. She says that they ¡’re looking for a metamodel of decision making and a general model for allowing multiattributes negotiations.
And…. that’s all, folks. Some questions and we’ll run for the next session. By, then.
Blogged with Flock
Bueno, por fin hemos llegado a Hammamet (Tünez) depués de un viaje accidentado: perdimos el enlace en Parías y tuvimos que esperar 2 horas en el aeropuesrto. El segundo avión salió casi una hora tarde y perdimos el autobús en Túnez. Allí nos tocó esperar 4 horas a que Mr. Transportation llegara con uno mayor.
Al final llegamos a Hammamet a las 9 de la noche. Me había levantado a las 4.30 de la mañana. Y, por supuesto, no llegué al 5th technical forum: la razón por la que vine a Túnez. Mola ¿eh?
Esta mañana ya nos hemos inscrito en la conferencia, hemos desayunado y ahora estoy en la conferencia invitada sobre el uso de técnicas de argumentación para negociación. Nos han dicho que teníamos que haber pagado el hotel a través de EUMAS. Imagino que eso supone que habrá que pagar un “extra” por algo (uso de las salas, WiFi o lo que sea), por eso era más caro.
Ya os contaré como va el resto.
Blogged with Flock
Una de las cosas que estamos mirando es cómo utilizar Android como plataforma de ejecución de agentes inteligentes. En principio, la idea es tener una arquitectura de agente que se pueda ejecutar sobre Android. Si además este agente es capaz de controlar el resto de aplicaciones nativas y acceder a sus datos, entonces sería fantástico y estaríamos cerca del agente que aparece en el vídeo Apple’s Knowledge Navigator (mov, 15 Mb).
Pero poco a poco.de momento, bastaría con que nuestro modelo de agente sea implementable en dispositivos móviles. Y la verdad es que los conceptos que se manejan en Andriod–actividades, intenciones, servicios y proveedores de contenidos–casan bastante bien con los componentes de nuestros agentes: comportamientos, capacidades, eventos y (por supuesto) creencias.
En principio, un agente está formado por un conjunto de comportamientos que se asocian a distintos escenarios o estados del mundo exterior. Cada comportamiento está formado por una serie de capacidades que establecen qué sabe o puede hacer el agente en cada situación. una capacidad representa una especie de regla Evento-Condición-Acción: cuando llega un evento determinado, si se cunple la condición de activación se ejecuta la acción asociada. Si un comportamiento está activo, todas sus capacidades están activas. Sólo se atenderá a los eventos que tengan alguna capacidad asociada, por lo que de esta forma se permite focalizar la atención del agente en los eventos que sean de interés o relevantes para el escenario en el que se encuentra actualmente.
La traducción de este esquema a entidades de Andriod es bastante intuitiva.
Blogged with Flock
Bueno, esto parece que ya casi está. Me falta arreglar todavía un par de cosillas, pero lo gordo está hecho. Ahora sólo falta empezar a rellenar esto con contenidos.
Por cierto, soy Miguel y este es mi nuevo blog…no, mi nueva web…, tampoco. Digamos que un poco de cada. Con esta remoledación espero conseguir un espacio más parcicipativo que una simple página web, aunque me va a dar penita dejar la vieja. Creo que me la voy a guardar y algún día podré una anotación de esas de “Evolución de mi web”. Por cierto, que tengo que buscar la primera. No sé si todavía podré rescatarla de algún sitio.
Puede que os encontréis trozos sin traducir, la parte de publicaciones es ridícula, me falta poner la licencia creative commons y no consigo escribir en la barra negra que tenéis arriba. Pero vamos, por lo demás, bien.
Miguel Rebollo Pedruelo
Dept. Stmas. Informáticos y Computación
Univ. Politécnica de Valencia
Camino de Vera s/n
46022 Valencia - España
Email: mrebollo [arroba] dsic.upv.es
Tfno: (+34) 963877000 ext 73584
Despacho: 2D31 (edif. DSIC)
Comentarios recientes