@phdthesis{Tan2023,
  author    = {Tan, Jing},
  title     = {Multi-Agent Reinforcement Learning for Interactive Decision-Making},
  doi       = {10.25932/publishup-60700},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-607000},
  school      = {Universit{\"a}t Potsdam},
  pages     = {xii, 135},
  year      = {2023},
  abstract  = {Distributed decision-making studies the choices made among a group of interactive and self-interested agents. Specifically, this thesis is concerned with the optimal sequence of choices an agent makes as it tries to maximize its achievement on one or multiple objectives in the dynamic environment. The optimization of distributed decision-making is important in many real-life applications, e.g., resource allocation (of products, energy, bandwidth, computing power, etc.) and robotics (heterogeneous agent cooperation on games or tasks), in various fields such as vehicular network, Internet of Things, smart grid, etc. This thesis proposes three multi-agent reinforcement learning algorithms combined with game-theoretic tools to study strategic interaction between decision makers, using resource allocation in vehicular network as an example. Specifically, the thesis designs an interaction mechanism based on second-price auction, incentivizes the agents to maximize multiple short-term and long-term, individual and system objectives, and simulates a dynamic environment with realistic mobility data to evaluate algorithm performance and study agent behavior. Theoretical results show that the mechanism has Nash equilibria, is a maximization of social welfare and Pareto optimal allocation of resources in a stationary environment. Empirical results show that in the dynamic environment, our proposed learning algorithms outperform state-of-the-art algorithms in single and multi-objective optimization, and demonstrate very good generalization property in significantly different environments. Specifically, with the long-term multi-objective learning algorithm, we demonstrate that by considering the long-term impact of decisions, as well as by incentivizing the agents with a system fairness reward, the agents achieve better results in both individual and system objectives, even when their objectives are private, randomized, and changing over time. Moreover, the agents show competitive behavior to maximize individual payoff when resource is scarce, and cooperative behavior in achieving a system objective when resource is abundant; they also learn the rules of the game, without prior knowledge, to overcome disadvantages in initial parameters (e.g., a lower budget). To address practicality concerns, the thesis also provides several computational performance improvement methods, and tests the algorithm in a single-board computer. Results show the feasibility of online training and inference in milliseconds. There are many potential future topics following this work. 1) The interaction mechanism can be modified into a double-auction, eliminating the auctioneer, resembling a completely distributed, ad hoc network; 2) the objectives are assumed to be independent in this thesis, there may be a more realistic assumption regarding correlation between objectives, such as a hierarchy of objectives; 3) current work limits information-sharing between agents, the setup befits applications with privacy requirements or sparse signaling; by allowing more information-sharing between the agents, the algorithms can be modified for more cooperative scenarios such as robotics.},
  language  = {en}
}
@phdthesis{Bazhenova2018,
  author    = {Bazhenova, Ekaterina},
  title     = {Discovery of Decision Models Complementary to Process Models},
  url       = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-410020},
  school      = {Universit{\"a}t Potsdam},
  year      = {2018},
  abstract  = {Business process management is an acknowledged asset for running an organization in a productive and sustainable way. One of the most important aspects of business process management, occurring on a daily basis at all levels, is decision making. In recent years, a number of decision management frameworks have appeared in addition to existing business process management systems. More recently, Decision Model and Notation (DMN) was developed by the OMG consortium with the aim of complementing the widely used Business Process Model and Notation (BPMN). One of the reasons for the emergence of DMN is the increasing interest in the evolving paradigm known as the separation of concerns. This paradigm states that modeling decisions complementary to processes reduces process complexity by externalizing decision logic from process models and importing it into a dedicated decision model. Such an approach increases the agility of model design and execution. This provides organizations with the flexibility to adapt to the ever increasing rapid and dynamic changes in the business ecosystem. The research gap, identified by us, is that the separation of concerns, recommended by DMN, prescribes the externalization of the decision logic of process models in one or more separate decision models, but it does not specify this can be achieved. The goal of this thesis is to overcome the presented gap by developing a framework for discovering decision models in a semi-automated way from information about existing process decision making. Thus, in this thesis we develop methodologies to extract decision models from: (1) control flow and data of process models that exist in enterprises; and (2) from event logs recorded by enterprise information systems, encapsulating day-to-day operations. Furthermore, we provide an extension of the methodologies to discover decision models from event logs enriched with fuzziness, a tool dealing with partial knowledge of the process execution information. All the proposed techniques are implemented and evaluated in case studies using real-life and synthetic process models and event logs. The evaluation of these case studies shows that the proposed methodologies provide valid and accurate output decision models that can serve as blueprints for executing decisions complementary to process models. Thus, these methodologies have applicability in the real world and they can be used, for example, for compliance checks, among other uses, which could improve the organization's decision making and hence it's overall performance.},
  language  = {en}
}