On Team Decision Problems with Nonclassical Information Structures

Malikopoulos, Andreas A.

doi:10.1109/tac.2022.3195126

Cited by 17 publications

(9 citation statements)

References 97 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our exposition, we restricted attention to centralized control systems. A potential future research direction includes expanding the framework to decentralized systems [37].…”

Section: Discussionmentioning

confidence: 99%

Separation of learning and control for cyber–physical systems

Malikopoulos

2023

Automatica

View full text Add to dashboard Cite

“…In our exposition, we restricted attention to centralized control systems. A potential future research direction includes expanding the framework to decentralized systems [37].…”

Section: Discussionmentioning

confidence: 99%

Separation of learning and control for cyber–physical systems

Malikopoulos

2023

Automatica

View full text Add to dashboard Cite

“…In a word-of-mouth communication structure, every member of the team communicates with her neighbors with delays in communication. This is a non-classical information structure [64] where the topological and temporal restrictions in communication mean that information propagates slowly through the team members.…”

Section: Properties Of the Expected Mobility Systemmentioning

confidence: 99%

A Control Framework for Socially-Optimal Emerging Mobility Systems

Malikopoulos¹

2023

Preprint

View full text Add to dashboard Cite

In this manuscript, we provide an online learning framework that will aim at distributing travel demand in a given transportation network resulting in a socially-optimal mobility system that travelers would be willing to accept. It is expected that connected and automated vehicles (CAVs) will gradually penetrate the market in ways that will improve safety and transportation efficiency over the next several years. However, we anticipate that efficient transportation and travel cost reduction might alter human travel behavior causing rebound effects, e.g., by improving effi-ciency, travel cost is decreased, hence willingness-to-travel is increased. The latter would increase overall vehicle miles traveled, which in turn might negate the benefits in terms of energy and travel time. Our aim is to develop a holistic and rigorous framework to capture the societal impact of CAVs and provide solutions that mitigate any potential rebound effects, e.g., increased vehicle miles traveled, increased travel demand, empty vehicle trips, while enhancing accessibility, safety, and equity in transportation.

show abstract

“…The most common modeling paradigm for such decision-making problems is the stochastic approach, where all disturbances to the system are considered to be random variables with known distributions, and the agent aims to select a decision-making strategy that minimizes the expected incurred cost [5]. Stochastic models have been utilized for problems in both control theory [6]- [13] and reinforcement learning [14]- [18]. A decision-making strategy derived using the stochastic approach performs optimally on average across numerous operations of the system.…”

Section: Introductionmentioning

confidence: 99%

Approximate Information States for Worst-Case Control and Learning in Uncertain Systems

Dave¹,

Nishanth²,

Malikopoulos³

2023

Preprint

View full text Add to dashboard Cite

In this paper, we investigate discrete-time decisionmaking problems in uncertain systems with partially observed states. We consider a non-stochastic model, where uncontrolled disturbances acting on the system take values in bounded sets with unknown distributions. We present a general framework for decision-making in such problems by developing the notions of information states and approximate information states. In our definition of an information state, we introduce conditions to identify for an uncertain variable sufficient to construct a dynamic program (DP) that computes an optimal strategy. We show that many information states from the literature on worstcase control actions, e.g., the conditional range, are examples of our more general definition. Next, we relax these conditions to define approximate information states using only output variables, which can be learned from output data without knowledge of system dynamics. We use this notion to formulate an approximate DP that yields a strategy with a bounded performance loss. Finally, we illustrate the application of our results in control and reinforcement learning using numerical examples.

show abstract

On Team Decision Problems with Nonclassical Information Structures

Cited by 17 publications

References 97 publications

Separation of learning and control for cyber–physical systems

Separation of learning and control for cyber–physical systems

A Control Framework for Socially-Optimal Emerging Mobility Systems

Approximate Information States for Worst-Case Control and Learning in Uncertain Systems

Contact Info

Product

Resources

About