scholarly journals A Sufficient Statistic for Influence in Structured Multiagent Environments

2021 ◽  
Vol 70 ◽  
pp. 789-870
Author(s):  
Frans Oliehoek ◽  
Stefan Witwicki ◽  
Leslie Kaelbling
Keyword(s):  
Decision Making ◽  
Single Agent ◽  
Sequential Decision ◽  
Sufficient Statistic ◽  
Compact Representations ◽  
Special Cases ◽  
Wide Range ◽  
Solution Methods ◽  
The One ◽  
Partially Observable

Making decisions in complex environments is a key challenge in artificial intelligence (AI). Situations involving multiple decision makers are particularly complex, leading to computational intractability of principled solution methods. A body of work in AI has tried to mitigate this problem by trying to distill interaction to its essence: how does the policy of one agent influence another agent? If we can find more compact representations of such influence, this can help us deal with the complexity, for instance by searching the space of influences rather than the space of policies. However, so far these notions of influence have been restricted in their applicability to special cases of interaction. In this paper we formalize influence-based abstraction (IBA), which facilitates the elimination of latent state factors without any loss in value, for a very general class of problems described as factored partially observable stochastic games (fPOSGs). On the one hand, this generalizes existing descriptions of influence, and thus can serve as the foundation for improvements in scalability and other insights in decision making in complex multiagent settings. On the other hand, since the presence of other agents can be seen as a generalization of single agent settings, our formulation of IBA also provides a sufficient statistic for decision making under abstraction for a single agent. We also give a detailed discussion of the relations to such previous works, identifying new insights and interpretations of these approaches. In these ways, this paper deepens our understanding of abstraction in a wide range of sequential decision making settings, providing the basis for new approaches and algorithms for a large class of problems.

scholarly journals A method for decision making with the OWA operator

2012 ◽  
Vol 9 (1) ◽  
pp. 357-380 ◽  
Author(s):  
José Merigó ◽  
Anna Gil-Lafuente
Keyword(s):  
Decision Making ◽  
Distance Measures ◽  
Weighted Averaging ◽  
Owa Operator ◽  
Minimum Level ◽  
Ordered Weighted Averaging ◽  
Special Cases ◽  
Wide Range ◽  
Decision Making Problem ◽  
Parameterized Family

A new method for decision making that uses the ordered weighted averaging (OWA) operator in the aggregation of the information is presented. It is used a concept that it is known in the literature as the index of maximum and minimum level (IMAM). This index is based on distance measures and other techniques that are useful for decision making. By using the OWA operator in the IMAM, we form a new aggregation operator that we call the ordered weighted averaging index of maximum and minimum level (OWAIMAM) operator. The main advantage is that it provides a parameterized family of aggregation operators between the minimum and the maximum and a wide range of special cases. Then, the decision maker may take decisions according to his degree of optimism and considering ideals in the decision process. A further extension of this approach is presented by using hybrid averages and Choquet integrals. We also develop an application of the new approach in a multi-person decision-making problem regarding the selection of strategies.

scholarly journals Induced Interval-Valued Intuitionistic Fuzzy Hybrid Aggregation Operators with TOPSIS Order-Inducing Variables

2012 ◽  
Vol 2012 ◽  
pp. 1-24 ◽  
Author(s):  
Jun-Ling Zhang ◽  
Xiao-Wen Qi
Keyword(s):  
Decision Making ◽  
Group Decision Making ◽  
Group Decision ◽  
Aggregation Operators ◽  
Intuitionistic Fuzzy ◽  
Multiple Attribute ◽  
Special Cases ◽  
Wide Range ◽  
Induced Aggregation ◽  
Interval Valued

Two induced aggregation operators with novelly designed TOPSIS order-inducing variables are proposed: Induced Interval-valued Intuitionistic Fuzzy Hybrid Averaging (I-IIFHA) operator and Induced Interval-valued Intuitionistic Fuzzy Hybrid Geometric (I-IIFHG) operator. The merit of two aggregation operators is that they can consider additional preference information of decision maker’s attitudinal characteristics besides argument-dependent information and argument-independent information. Some desirable properties of I-IIFHA and I-IIFHG are studied and theoretical analysis also shows that they can include a wide range of aggregation operators as special cases. Further, we extend these operators to form a novel group decision-making method for selecting the most desirable alternative in multiple attribute multi-interest group decision-making problems with attribute values and decision maker’s interest values taking the form of interval-valued intuitionistic fuzzy numbers, and application research to real estate purchase selection shows its practicality.

scholarly journals Multicriteria pathfinding in uncertain simulated environments

2018 ◽  
Author(s):  
◽  
Andrew R. Buck
Keyword(s):  
Decision Making ◽  
Decision Maker ◽  
Multicriteria Decision Making ◽  
Spatial Knowledge ◽  
Sequential Decision ◽  
Multicriteria Decision ◽  
Multicriteria Decision Making Problems ◽  
Trade Offs ◽  
Mental Map ◽  
Partially Observable

Multicriteria decision-making problems arise in all aspects of daily life and form the basis upon which high-level models of thought and behavior are built. These problems present various alternatives to a decision-maker, who must evaluate the trade-offs between each one and choose a course of action. In a sequential decision-making problem, each choice can influence which alternatives are available for subsequent actions, requiring the decision-maker to plan ahead in order to satisfy a set of objectives. These problems become more difficult, but more realistic, when information is restricted, either through partial observability or by approximate representations. Pathfinding in partially observable environments is one significant context in which a decision-making agent must develop a plan of action that satisfies multiple criteria. In general, the partially observable multiobjective pathfinding problem requires an agent to navigate to certain goal locations in an environment with various attributes that may be partially hidden, while minimizing a set of objective functions. To solve these types of problems, we create agent models based on the concept of a mental map that represents the agent's most recent spatial knowledge of the environment, using fuzzy numbers to represent uncertainty. We develop a simulation framework that facilitates the creation and deployment of a wide variety of environment types, problem definitions, and agent models. This computational mental map (CMM) framework is shown to be suitable for studying various types of sequential multicriteria decision-making problems, such as the shortest path problem, the traveling salesman problem, and the traveling purchaser problem in multiobjective and partially observable configurations.

scholarly journals A Novel Heterogeneous Swarm Reinforcement Learning Method for Sequential Decision Making Problems

2019 ◽  
Vol 1 (2) ◽  
pp. 590-610
Author(s):  
Zohreh Akbari ◽  
Rainer Unland
Keyword(s):  
Decision Making ◽  
Reinforcement Learning ◽  
Single Agent ◽  
Sequential Decision Making ◽  
Multi Agent Systems ◽  
Sequential Decision ◽  
Agent Systems ◽  
Novel Approach ◽  
Markov Decision ◽  
Multi Agent

Sequential Decision Making Problems (SDMPs) that can be modeled as Markov Decision Processes can be solved using methods that combine Dynamic Programming (DP) and Reinforcement Learning (RL). Depending on the problem scenarios and the available Decision Makers (DMs), such RL algorithms may be designed for single-agent systems or multi-agent systems that either consist of agents with individual goals and decision making capabilities, which are influenced by other agent’s decisions, or behave as a swarm of agents that collaboratively learn a single objective. Many studies have been conducted in this area; however, when concentrating on available swarm RL algorithms, one obtains a clear view of the areas that still require attention. Most of the studies in this area focus on homogeneous swarms and so far, systems introduced as Heterogeneous Swarms (HetSs) merely include very few, i.e., two or three sub-swarms of homogeneous agents, which either, according to their capabilities, deal with a specific sub-problem of the general problem or exhibit different behaviors in order to reduce the risk of bias. This study introduces a novel approach that allows agents, which are originally designed to solve different problems and hence have higher degrees of heterogeneity, to behave as a swarm when addressing identical sub-problems. In fact, the affinity between two agents, which measures the compatibility of agents to work together towards solving a specific sub-problem, is used in designing a Heterogeneous Swarm RL (HetSRL) algorithm that allows HetSs to solve the intended SDMPs.

scholarly journals Linguistic Weighted Aggregation under Confidence Levels

2015 ◽  
Vol 2015 ◽  
pp. 1-7
Author(s):  
Chonghui Zhang ◽  
Weihua Su ◽  
Shouzhen Zeng ◽  
Linyun Zhang
Keyword(s):  
Decision Making ◽  
Multicriteria Decision Making ◽  
Aggregation Operators ◽  
Weighted Averaging ◽  
New Approach ◽  
Confidence Levels ◽  
Ordered Weighted Averaging ◽  
Special Cases ◽  
Wide Range ◽  
Information Providers

We develop some new linguistic aggregation operators based on confidence levels. Firstly, we introduce the confidence linguistic weighted averaging (CLWA) operator and the confidence linguistic ordered weighted averaging (CLOWA) operator. These two new linguistic aggregation operators are able to consider the confidence level of the aggregated arguments provided by the information providers. We also study some of their properties. Then, based on the generalized means, we introduce the confidence generalized linguistic ordered weighted averaging (CGLOWA) operator. The main advantage of the CGLOWA operator is that it includes a wide range of special cases such as the CLOWA operator, the confidence linguistic ordered weighted quadratic averaging (CLOWQA) operator, and the confidence linguistic ordered weighted geometric (CLOWG) operator. Finally, we develop an application of the new approach in a multicriteria decision-making under linguistic environment and illustrate it with a numerical example.

scholarly journals Responsibility Under Uncertainty: Which Climate Decisions Matter Most?

2021 ◽  
Author(s):  
Nicola Botta ◽  
Nuria Brede ◽  
Michel Crucifix ◽  
Cezar Ionescu ◽  
Patrik Jansson ◽  
...  
Keyword(s):  
Decision Making ◽  
Decision Maker ◽  
Global Economy ◽  
Decision Processes ◽  
Sequential Decision ◽  
Causal Relevance ◽  
Natural Conditions ◽  
Wide Range ◽  
Probability Of Success ◽  
The Right

Abstract We propose a new method for estimating how much decisions under monadic uncertainty matter. The method is generic and suitable for measuring responsibility in finite horizon sequential decision processes. It fulfills “fairness” requirements and three natural conditions for responsibility measures: agency, avoidance and causal relevance. We apply the method to study how much decisions matter in a stylized greenhouse gas emissions process in which a decision maker repeatedly faces two options: start a “green” transition to a decarbonized society or further delay such a transition. We account for the fact that climate decisions are rarely implemented with certainty and that their consequences on the climate and on the global economy are uncertain. We discover that a “moral” approach towards decision making – doing the right thing even though the probability of success becomes increasingly small – is rational over a wide range of uncertainties.

An Introduction to Fully and Partially Observable Markov Decision Processes

2012 ◽  
pp. 33-62 ◽  
Author(s):  
Pascal Poupart
Keyword(s):  
Decision Making ◽  
Markov Decision Processes ◽  
Decision Processes ◽  
Sequential Decision Making ◽  
Decision Making Under Uncertainty ◽  
Sequential Decision ◽  
Markov Decision ◽  
The Common ◽  
Partially Observable Markov ◽  
Partially Observable

The goal of this chapter is to provide an introduction to Markov decision processes as a framework for sequential decision making under uncertainty. The aim of this introduction is to provide practitioners with a basic understanding of the common modeling and solution techniques. Hence, we will not delve into the details of the most recent algorithms, but rather focus on the main concepts and the issues that impact deployment in practice. More precisely, we will review fully and partially observable Markov decision processes, describe basic algorithms to find good policies and discuss modeling/computational issues that arise in practice.

Sign in / Sign up

Export Citation Format

Share Document