Spelling suggestions: "subject:"8ecision process."" "subject:"becision process.""
121 |
Designförslag på belöningsfunktioner för självkörande bilar i TORCS som inte krockar / Design suggestion on reward functions for self-driving cars in TORCS that do not crashAndersson, Björn, Eriksson, Felix January 2018 (has links)
Den här studien använder sig av TORCS (The Open Racing Car Simulator) som är ett intressant spel att skapa självkörande bilar i då det finns nitton olika typer av sensorer som beskriver omgivningen för agenten. Problemet för denna studie har varit att identifiera vilka av alla dessa sensorer som kan användas i en belöningsfunktion och hur denna sedan skall implementeras. Studien har anammat en kvantitativa experimentell studie där forskningsfrågan är: Hur kan en belöningsfunktion utformas så att agenten klarar av att manövrera i spelet TORCS utan att krocka och med ett konsekvent resultat Den kvantitativ experimentell studien valdes då författarna behövde designa, implementera, utföra experiment och utvärdera resultatet för respektive belöningsfunktion. Det har utförts totalt femton experiment över tolv olika belöningsfunktioner i spelet TORCS på två olika banor E-Track 5(E-5) och Aalborg. De tolv belöningsfunktionerna utförde varsitt experiment på E-5 där de tre som fick bäst resultat: Charlie, Foxtrot och Juliette utförde ett experiment på Aalborg, då denna är en svårare bana. Detta för att kunna styrka om den kan köra på mer än en bana och om belöningsfunktionen då är generell. Juliette är den belöningsfunktion som var ensam med att klara både E-5 och Aalborg utan att krocka. Genom de utförda experimenten drogs slutsatsen att Juliette uppfyller forskningsfrågan då den klarar bägge banorna utan att krocka och när den lyckas får den ett konsekvent resultat. Studien har därför lyckats designa och implementera en belöningsfunktion som uppfyller forskningsfrågan. / For this study TORCS (The Open Racing Car Simulator) have been used, since it is an interesting game to create self-driving cars in. This is due to the fact there is nineteen different sensors available that describes the environment for the agent. The problem for this study has been to identify what sensor can be used in a reward function and how should this reward function be implemented. The study have been utilizing a quantitative experimental method where the research questions have been: How can a reward function be designed so that an Agent can maneuver in TORCS without crashing and at the same time have a consistent result The quantitative experimental method was picked since the writer’s hade to design, implement, conduct experiment and evaluate the result for each reward function. Fifteen experiments have been conducted over twelve reward functions on two different maps: E-Track 5 (E-5) and Aalborg. Each of the twelve reward function conducted an experiment on E-5, where the three once with the best result: Charlie, Foxtrot and Juliette conducted an additional experiment on Aalborg. The test on Aalborg was conducted in order to prove if the reward function can maneuver on more than one map. Juliette was the only reward function that managed to complete a lap on both E-5 and Aalborg without crashing. Based on the conducted experiment the conclusion that Juliette fulfills the research question was made, due to it being capable of completing both maps without crashing and if it succeeded it gets a consistent result. Therefor this study has succeeded in answering the research question.
|
122 |
Optimal mobility patterns in epidemic networksNirkhiwale, Supriya January 1900 (has links)
Master of Science / Department of Electrical and Computer Engineering / Caterina M. Scoglio / Disruption Tolerant Networks or opportunistic networks represent a class of networks where there is no contemporaneous path from source to destination. In other words, these are networks with intermittent connections. These networks are generally sparse or highly mobile wireless networks. Each node has a limited radio range and the connections between nodes may be disrupted due to node movement, hostile environments or power sleep schedules, etc. A common example of such networks is a sensor network monitoring nature or military field or a herd of animals under study. Epidemic routing is a widely proposed routing mechanism for data propagation in these type of networks. According to this mechanism, the source copies its packets to all the nodes it meets in its radio range. These nodes in turn copy the received packets to the other nodes they meet and so on. The data to be transmitted travels in a way analogous to the spread of an infection in a biological network. The destination finally receives the packet and measures are taken to eradicate the packet from the network. The task of routing in epidemic networks faces certain difficulties involving minimizing the delivery delay with a reduced consumption of resources. Every node has severe power constraints and the network is also susceptible to temporary but random failure of nodes. In the previous work, the parameter of mobility has been considered a constant for a certain setting. In our setting, we consider a varying parameter of mobility. In this framework, we determine the optimal mobility pattern and a forwarding policy that a network should follow in order to meet the trade-off between delivery delay and power consumption. In addition, the mobility pattern should be such that it can be practically incorporated. In our work, we formulate an optimization problem which is solved by using the principles of dynamic programming. We have tested the optimal algorithm through extensive simulations and they show that this optimization problem has a global solution.
|
123 |
AYE, AYE, CAPTAIN : En undersökning av ansvar och beslutsprocesser inom sjöfartNorberg, Peter January 2013 (has links)
Beslutsfattande är den viktigaste av alla ledningsaktiviteter. Syftet med denna studie är att undersöka hur befälhavare upplever beslutprocesserna ombord och i den omgivande organisationen. Undersökningen har gjorts genom intervjuer med tre befälhavare. Intervjuer valdes för att kunna lyssna till intervjupersonernas upplevda värld och de har varit semistrukturerade, som stöd har en intervjuguide använts. Det går i den här formen att följa trådar eller infall men också att gå tillbaka och förtydliga. Som underlag för analysen beskrivs teorier om beslutsprocesser, ansvar inom sjöfarten och arbete i team eller grupper. Analysen har sedan gjorts utifrån en modell i beslutsteorin. Resultatet har blivit en berättelse genom beslutsprocessen och den har givit två intressanta slutsatser nämligen: att när styrning från managementbolag allt mer börjar användas upplever sig befälhavarna maktlösa och långt från besluten fast de å andra sidan är tydliga med att de bestämmer och skaffar sig strategier för att behålla beslutet. / Decision making is the most important of all managerial activities. The purpose of this study is to examine how commanders experience decision-making processes on board and in the surrounding organization. The survey was conducted through interviews with three masters. Interviews were chosen to listen to the interviewees’ perceived world and they have been semi-structured, as support an interview guide has been used. In this shape it is possible to follow threads or whim but also to go back and clarify. As a background for analyze, theories of decision-making, responsibility in shipping and work in teams or groups are described. The analysis has been based on a model of decision theory. The result is a story by the decision making process and it gave two interesting conclusions: when control by management companies is increasing masters experience loss of power and distance to decisions but on the other hand they are clear that they are in charge and acquire strategies for retaining that.
|
124 |
Faktory ovlivňující nákupní chování ve vztahu ke koupi automobilu / Factors influencing purchase behavior in relation to automobile purchaseKlementová, Jana January 2014 (has links)
| 1 Abstract The aim of the thesis is the contentual and procedural analysis of customers purchase behavior. Both analyses point out several internal and external factors, which play their role in customer's decision making process. The result of this process is or is not the purchase of the product. The combined research enables to identify the factors and determine their importance. In the qualitative research part parameters of the research are set based on semi-structured interviews. The parameters are then quantitatively measured according to their importance, after that they are analysed by means of factor analysis and via comparison the statistically important differences in decision making process among several target groups. Considering also the external factors influencing the sales process the perception of the same parameters by salesmen is researched as well, because they are a part of the customers decision making process and whose expectations also actively affect the process. The discovered factors were then compared with the Maslow's hierarchy of needs. From the other results of the research the conclusion was drawn that there do exist statistically important gender differences, differences between perception of the customers needs among various owners of sales points and among salesmen....
|
125 |
Informationens effekt i moraliska dilemman : Effekterna mängden tillgänglig information har på spelares beslut i prohibition dilemman / The role of information in moral dilemmas : The effects the amount of available information has on player decisions in prohibition dilemmasPethrus, Anton January 2017 (has links)
Detta arbete redogör för effekterna tillgången eller avsaknaden av omfattande karaktärsinformation samt information om beslutens direkta konsekvenser har på spelarens beslutsprocess i narrativa prohibition dilemman. Då det finns avsaknad av konsensus om dess effekter i ämnesområdet och en frånvaro av objektiva studier för att undersöka de aktuella ståndpunkterna, är studien nödvändig för att förbättra kunskapen inom området och för att generera objektiva resultat av dess effekter på spelare. För att undersöka problemformuleringen skapades två likadana textbaserade spel, där skillnaden mellan dem var att den ena gav spelaren omfattande karaktärsinformation och information om beslutens direkta konsekvenser i spelet narrativa prohibition dilemman, medan den andra artefakten saknade den informationen. Respektive artefakt testades på en separat testgrupp för att generera resultat om dess respektive effekt på testgruppen som spelade den. Resultaten var blandade i de olika testgrupperna. Det gick dock att finna tendenser i resultaten i de olika grupperna. När spelare inte hade omfattande karaktärsinformation samt information om beslutens konsekvenser, var de mer benägna att välja de beslut de kände var rätt utifrån deras egna värderingar när de inte kunde härleda beslutens konsekvenser. När de kunde härleda beslutens konsekvenser valde de ofta det val vilket gav dem den mest fördelaktiga utkomsten. När spelaren hade tillgång omfattande karaktärsinformation samt information om beslutens direkta konsekvenser motiverades deras beslut av att väga fördelarna och nackdelarna av respektive beslutsalternativ mot varandra, utifrån en strategisk och/eller moralisk basis, där de tog det alternativ som gav dem den mest optimala utkomsten. I fall där spelarna hade en emotionell investering i besluten valde de oftast oavsett mängden tillgänglig information vid beslutet att välja det beslut som kändes rätt. Något som även framkom i studien var att bättre resultat för att besvara problemformuleringen och explorativt undersöka ämnesområdet, skulle vara att göra ett liknande test av artefakterna på grupper av definierade spelartyper. Det skulle ge en bättre översikt av respektive utgångspunkts påverkan på specifika spelartyper och generera tydligare samt mer användbara resultat att besvara problemformuleringen med.
|
126 |
Exploring the customer journey : An exploratory study investigating the customer journeyBolinder, Marcus, Philip, Boström January 2019 (has links)
The customer journey is a re-submerged subject which has become increasingly relevant. The focus derives from the increased focus on customer experience as touch points are increasing as well as becoming more accessible. This is creating problems for companies to allocate their resources in marketing. Previous literature on customer journeys are also limited, mostly originating from research on customer experience. This motivates research within the field with the purpose of exploring the customer journey and connected phenomena. The basis of this thesis was the customer journey which derives from customer experience. Further, the theoretical framework presents theories about customer experience, consumer buying decisions, customer journey and its touch points. The theoretical framework also presents two conceptual models concerning customer journey and customer experience. Focus was here on exploring the formulation of the customer journey, its steps and components. But also, on investigating how customer experiences affects customer journeys. This thesis had a qualitative research strategy and empirical material was collected through interviews. The empirical findings and the analysis resulted in several conclusions as this was an exploratory study. Reason for use of route, use of different touch points, customer experience impact and were found. Furthermore, the customer journey itself was identified as a reason for conducting a purchase. The result of this thesis might help companies allocate their resources more efficiently between touch points. As well as understanding how to create positive customer experience and the importance of it.
|
127 |
Får det lov att vara en bil? : En kvalitativ studie om köpbeslutsprocessen inom bilhandeln och hur det fysiska säljmötet har förändrats. / May I offer you a car? : A qualitative study on the purchasing decision process in car sales and how the physical sales meeting has changed.Engström, Sofie, Mattsson, Pontus, Olofsson, Philip January 2019 (has links)
Bakgrund: Studien har undersökt hur köpbeslutsprocessen och det fysiska mötet ser ut i bilhandeln samt om det finns skillnader mellan hur kvinnor och män upplever köpbeslutsprocessen inom bilhandeln. Studiens ämne valdes på grund av att det gjorts tidigare studier inom bilhandeln gällande köpbeslutsprocessen, men att det finns en brist sedan det digitala mötet kom in i bilden. Det har även genomförts flera studier om skillnader mellan mäns och kvinnors skillnader i köpbeteende, men det finns inga som studerar om skillnaderna ser likadana ut i bilbranschen. Forskningsfråga: Studiens huvudfråga lyder: Hur interagerar bilhandeln med kunden i köpbeslutsprocessen genom det fysiska säljmötet för att det ska leda till försäljning av bil? Studiens underfråga lyder: Skiljer sig bilköp åt för män och kvinnor och i sådana fall på vilket sätt? Syfte: Syftet med studien är att undersöka hur bilhandeln kan arbeta vid köpbeslutsprocessen i förhållande till det fysiska säljmötet och vad som gör att kunden genomför ett köp av bil.Vidare är syftet att studera om det finns några skillnader mellan män och kvinnor vid köpet av bil och om det finns, på vilket sätt skiljer de sig. Metod: Den här studien har en kvalitativ undersökningsmetod och ett abduktivt angreppssätt. Resultatet i studien bygger på 12 semistrukturerade intervjuer med respondenter spridda över Sverige. Slutsats: Studien har visat att köpbeslutsprocessen är oförändrad, i sin struktur, av den digitala närvaron. Däremot ser det fysiska mötet annorlunda ut då säljaren behöver inleda med att ta reda på hur långt kunden kommit i sin köpbeslutsprocess. Studiens underfråga har medfört resultat som visar att kvinnors och mäns evolutionära roller byter plats inom bilindustrin då det är män som uppskattar att handla mer hedonistiskt och kvinnor mer funktionellt. / Background: The study has examined how the buying decision process and the physical meeting work in car sales and whether there are differences between how women and men experience the buying decision process in the industry. The subject of the study was chosen due to the fact that there are previous studies in car sales regarding the buying decision process, but that there has been a shortage since the digital meeting came into the picture. There have also been several studies on differences in buying behaviour between men and women, but there are no studies regarding whether the differences look the same in the car industry. Research question: The main question of the study is: How does the car sales interact with the customers in the buying decision process through the physical sales meeting in order for it to lead to sales of a car? The study´s sub question reads: Do car purchase differ for men and women, and in such a case how? Purpose: The purpose of the study is to investigate how car sales can work in the buying decision process in relation to the physical sales meeting and what makes the customer to purchase a car. Furthermore, the aim is to study whether there are any differences between men and women in the purchase of a car and, if there is, how they differ. Method: This study has applied a qualitative method and abductive approach. The results of the study are based on 12 semi-structured interviews with respondents spread across Sweden. Conclusion: The study has shown that the buying decision process is unchanged, in its structure, of the digital presence. However, the physical meeting looks different as the seller needs to start by finding out how far the customer has come in the buying decision process. The study's sub question has resulted in results that show that women and men's evolutionary roles change places in the automotive industry as it is the men who appreciate to buy more hedonic and women more functionally.
|
128 |
Prefetching control for on-demand contents distribution : a Markov decision process study / Contrôle du préchargement pour la distribution de contenus à la demande : une approche par les processus de décision markoviensMorad, Olivia 17 September 2014 (has links)
Le contexte de la thèse porte sur le contrôle des réseaux de distribution de contenu à la demande. La performance des systèmes distribués interactifs dépend essentiellement sur la prévision du comportement de l'utilisateur et la bande passante en tant que ressource de réseau critique. Le préchargement est une approche prédictive bien connu dans le World Wide Web ce qui évite les délais de réponse en exploitant un temps d'arrêt que permet d'anticiper les futures demandes de l'utilisateur et prend avantage des ressources réseau disponibles. Le contrôle de préchargement est une opération vitale pour les systèmes à la demande interactifs où la réponse instantanée est le facteur crucial pour la réussite du système. Le contrôleur en ce type de système interactif fonctionne dans un environnement incertain et rend séquences de décisions à court et long terme effets stochastique. La difficulté est alors de déterminer à chaque état du système les contenus préchargés dans le cache. Le plan de préchargement pendant une session en flux continu interactif peut être modélisé comme un problème de décision séquentielle par les processus de décision de Markov (MDP). Nous nous concentrons sur le problème de contrôle de préchargement, dans lequel le contrôleur cherche à atteindre l'état du système à coût zéro aussi vite que possible. Nous modélisons ce problème de contrôle comme un problème de programmation dynamique stochastique négatif dans lequel nous minimisons le coût total prévu. Dans ce contexte, nous avons abordé les questions de recherche suivantes: 1) Comment fournir un politique de préchargement optimale/ approximative optimale qui maximise l'utilisation de la bande passante tout en minimisant les coûts de blocage et de la latence de l'utilisateur engagés sur le chemin? 2) Comment exploiter la structure du modèle de contrôle de préchargement pour aider efficacement calculer la politique de contrôle de préchargement avec la réduction des efforts de calcul et la mémoire de stockage? 3) Comment mener une étude d'évaluation pour évaluer le préchargement de différents algorithmes heuristiques basée sur le contexte de l'optimisation au lieu du cadre de l'empirique / simulation. Pour l'étude de notre problème de recherche, nous avons développé notre modèle MDP de préchargement, PREF-CT, nous avons établi ses propriétés théoriques et nous avons résolu par l'algorithme Value Iteration comme algorithme MDP pour calculer la politique de préchargement optimale. Pour calcul de la politique de préchargement optimale efficace, nous avons détecté une structure spéciale qui réalise un modèle de contrôle plus compact. Cette structure spéciale permet de développer deux algorithmes différents stratégiquement qui améliorent la complexité du calcul de la politique de préchargement optimale: - la première est « ONE-PASS » le second est « TREE-DEC ». Pour surmonter le problème de la dimensionnalité résultant du calcul de la politique de préchargement optimale, nous avons proposé l'algorithme de préchargement heuristique: « Relevant Blocks Prefetching » (RBP). Pour évaluer et comparer le préchargement politiques calculés par des algorithmes de préchargement heuristiques différents, nous avons présenté un cadre fondé sur des différentes mesures de performance. Nous avons appliqué le cadre proposé sous différentes configurations de coûts et différents comportements des utilisateurs pour évaluer les politiques de préchargement calculées par notre algorithme de préchargement proposé; RBP. Par rapport aux politiques de préchargement optimales, l'analyse expérimentale a prouvé des performances significatives des politiques de préchargement de l'heuristique du RBP algorithme. En outre, l'algorithme heuristique de préchargement; RBP se distingue par une propriété de clustériser qui est important pour réduire considérablement la mémoire nécessaire pour stocker la politique de préchargement. / The thesis context is concerned with the control of theOn-demand contents distribution networks. The performance of suchinteractive distributed systems basically depends on the prediction ofthe user behavior and the bandwidth as a critical network resource.Prefetching is a well-known predictive approach in the World Wide Webwhich avoids the response delays by exploiting some downtime thatpermits to anticipate the user future requests and takes advantage ofthe available network resources. Prefetching control is a vitaloperation for the On-demand interactive systems where the instantaneousresponse is the crucial factor for the system success. The controller insuch type of interactive system operates in an uncertain environment andmakes sequences of decisions with long and short term stochasticeffects. The difficulty, then, is to determine at every system statewhich contents to prefetch into the cache. The prefetching plan duringan interactive streaming session can be modeled as a sequential decisionmaking problem by a Markov Decision Process (MDP). We focus on theprefetching control problem in which the controller seeks to reach aZero-Cost system state as quickly as possible. We model this controlproblem as a Negative Stochastic Dynamic Programming problem in which weminimize the undiscounted total expected cost. Within this context, weaddressed the following research questions: 1) How to provide anoptimal/approximate-optimal prefetching policy that, maximizes thebandwidth utilization while minimizes the user's blocking and latencycosts incurred along the way? 2) How to exploit structure in theprefetching control model to help efficiently compute such prefetchingcontrol policy with both computational efforts and storage memoryreduction? 3) How to conduct a performance evaluation study to evaluatedifferent prefetching heuristic algorithms based on the context of thecontrol optimization rather than the context of theempirical/simulation. For studying our research problem, we developedour MDP prefetching control model, PREF-CT, we established itstheoretical properties and we solved it by the Value Iteration algorithmas MDP algorithm for computing the optimal prefetching policy. Forcomputing the optimal prefetching policy efficiently, we detected aspecial structure that achieves more compact control model. This specialstructure permits to develop two strategically different algorithmswhich improve the complexities of computing the optimal prefetchingpolicy: - the first one is the ONE-PASS which is based mainly on solvinga system of linear equations simultaneously in only one iteration,whereas the second is the TREE-DEC which is based on Markov decisiontree decomposition in which sequential sets of systems of equations aresolved. For overcoming the problem of the curse of dimensionalityresulting from the computation of the optimal prefetching policy, weproposed the prefetching heuristic algorithm: the Relevant BlocksPrefetching algorithm (RBP). For evaluating and comparing prefetchingpolicies computed by different prefetching heuristic algorithms, wepresented a framework based on different performance measures. Weapplied the suggested framework under different costs configurations anddifferent user behaviors to evaluate the prefetching policies computedby our proposed prefetching heuristic algorithm; the RBP. Compared tothe optimal prefetching policies, the experimental analysis provedsignificant performance of the prefetching policies of the RBP heuristicalgorithm. In addition, the RBP prefetching heuristic algorithm isdistinguished by a clustering property which is of importance to reducesignificantly the memory necessary to store the prefetching policy tothe controller.
|
129 |
Immateriella investeringsbeslut : En kvalitativ studie om beslutsprocessen vid immateriella investeringar och faktorer vid immateriella investeringsbeslut / Intangible investment decisions : A qualitative study on the decision-making process for intangible investments and factors in intangible investment decisionsOlin, Anna, Johnsson, Julia January 2019 (has links)
Bakgrund och problem: Immateriella investeringar har ökat inom det svenska näringslivet och utgör en betydande del av landets totala investeringar. Fortsättningsvis är immateriella investeringar väsentliga för samhällets och företagens produktivitet samt konkurrenskraft. Utvecklingen av immateriella investeringar tyder på att betydelsen av immateriella investeringsbeslut ökat. Trots det uppmärksammas ämnesområdet inte av den nuvarande forskningen eller litteraturen i en större omfattning. Syfte: Syftet är att studien ska bidra till en djupare förståelse angående immateriella investeringar genom att generera kunskap inom ett område som inte tidigare uppmärksammats i en större omfattning. Uppsatsen avser att tillföra information i relation till den immateriella beslutsprocessen samt angående faktorer som anses utgöra en väsentlig påverkan på det immateriella investeringsbeslutet. Metod: Studien avser en kvalitativ flerfallsdesign där fyra företag har utgjort det empiriska underlaget. Vidare genomfördes intervjuerna på ett semi-strukturerat tillvägagångssätt för att erhålla information.Slutsats: De immateriella investeringarnas beslutsprocessen initieras genom en identifiering av problem och målsättning, om företaget har upplevt ett initialt dilemma. Därefter framställs olika investeringsalternativ som utvärderas utifrån väsentliga faktorer. Avslutningsvis genomförs det immateriella investeringsbeslutet av högre insatser i organisationen. Väsentliga faktorer vid det immateriella investeringsbeslutet är lönsamhet, ny teknologi, tillväxt och riskbedömning. Nyckelord: Immateriella investeringar, investeringsbeslut, beslutsunderlag, immateriell beslutsprocess / Background and problem: Intangible investments have increased in the Swedish business sector and constitute a significant part of the country's total investments. Intangible investments are essential for the society's productivity and companies competitiveness. The development of intangible investments indicates that the importance of intangible investment decisions has increased. Despite this, the subject area is not noticed by the current research or literature to a greater extent. Purpose: The purpose of this study is that it will contribute to a deeper understanding of intangible investments by generating knowledge in an area that has not previously been recognized to a greater extent. The paper intends to provide information in relation to the intangible decision-making process and regarding factors that are considered to have an impact on the intangible investment decision. Method: The thesis is based on a qualitative multi-case design in which four companies have constituted to the empirical basis. Furthermore, the interviews were conducted on a semi-structured approach to obtaining information.Conclusion: The decision-making process of the intangible investments is initiated through an identification of problems and goals, if the company has experienced a first dilemma. Then there are different investment options that are evaluated based on essential factors. Finally, the intangible investment decision is made by higher units in the organization. Key factors in the intangible investment decision are profitability, new technology, growth and risk assessment. Keywords: Intangible investments, investment decisions, decision basis, intangible decision process
|
130 |
Estratégias para otimização do algoritmo de Iteração de Valor Sensível a Risco / Strategies for optimization of Risk Sensitive Value Iteration algorithmIgor Oliveira Borges 11 October 2018 (has links)
Processos de decisão markovianos sensíveis a risco (Risk Sensitive Markov Decision Process - RS-MDP) permitem modelar atitudes de aversão e propensão ao risco no processo de tomada de decisão usando um fator de risco para representar a atitude ao risco. Para esse modelo, existem operadores que são baseados em funções de transformação linear por partes que incluem fator de risco e fator de desconto. Nesta dissertação são formulados dois algoritmos de Iteração de Valor Sensível a Risco baseados em um desses operadores, esses algoritmos são chamados de Iteração de Valor Sensível a Risco Síncrono (Risk Sensitive Value Iteration - RSVI) e Iteração de Valor Sensível a Risco Assíncrono (Asynchronous Risk Sensitive Value Iteration- A-RSVI). Também são propostas duas heurísticas que podem ser utilizadas para inicializar os valores dos algoritmos de forma a torná-los mais eficentes. Os resultados dos experimentos no domínio de Travessia do Rio em dois cenários de recompensas distintos mostram que: (i) o custo de processamento de políticas extremas a risco, tanto de aversão quanto de propensão, é elevado; (ii) um desconto elevado aumenta o tempo de convergência do algoritmo e reforça a sensibilidade ao risco adotada; (iii) políticas com valores para o fator de risco intermediários possuem custo computacional baixo e já possuem certa sensibilidade ao risco dependendo do fator de desconto utilizado; e (iv) o algoritmo A-RSVI com a heurística baseada no fator de risco pode reduzir o tempo para o algoritmo convergir, especialmente para valores extremos do fator de risco / Risk Sensitive Markov Decision Process (RS-MDP) allows modeling risk-averse and risk-prone attitudes in decision-making process using a risk factor to represent the risk-attitude. For this model, there are operators that are based on a piecewise linear transformation function that includes a risk factor and a discount factor. In this dissertation we formulate two Risk Sensitive Value Iteration algorithms based on one of these operators, these algorithms are called Synchronous Risk Sensitive Value Iteration (RSVI) and Asynchronous Risk Sensitive Value Iteration (A-RSVI). We also propose two heuristics that can be used to initialize the value of the RSVI or A-RSVI algorithms in order to make them more efficient. The results of experiments with the River domain in two distinct rewards scenarios show that: (i) the processing cost in extreme risk policies, for both risk-averse and risk-prone, is high; (ii) a high discount value increases the convergence time and reinforces the chosen risk attitude; (iii) policies with intermediate risk factor values have a low computational cost and show a certain sensitivity to risk based on the discount factor; and (iv) the A-RSVI algorithm with the heuristic based on the risk factor can decrease the convergence time of the algorithm, especially when we need a solution for extreme values of the risk factor
|
Page generated in 0.063 seconds