Artificial intelligence (AI) and especially reinforcement learning (RL) have the potential to enable agents learn perform tasks autonomously with superhuman performance. However, we consider RL as fundamentally a Human-in-the-Loop (HITL) paradigm, even when an agent eventually performs its task autonomously. In cases where reward function is challenging or impossible define, HITL approaches are considered particularly advantageous. The application of Reinforcement Learning from Human...

Recently, a model of decentralized control system with local and remote controllers connected over unreliable channels was presented in [1]. The has nonclassical information structure that is not partially nested. Nonetheless, it shown [1] the optimal strategies are linear functions state estimate (which nonlinear function observations). Their proof based on fairly sophisticated dynamic programming argument. In this article, we present an alternative elementary result which uses common...

Abstract In this paper, the effects of Maxwell nanoliquid transmission with MWCNT nanotube are investigated over two circular wires opposite currents and a stretching sheet. The innovation paper is examination temperature nanofluid velocity also magnetic potential effect in three different places on finite element method utilized for calculating differential equations. By increasing distance from sheet, amount increased, large vortex forms at bottom Overall, around beginning sheet 46.09%...

Solid organ transplantation patients are at high risk for opportunistic air-borne fungal infections due to using the potent immunosuppressive agents.The current study aimed qualitatively and quantitatively evaluate flora present in air of Kidney transplant unit Baqiyatallah hospital.In this prospective study, samples from patient room, baths site, ICU isolated corridor site outside ward were obtained by settled plate technique plates containing Sabouraud's dextrose agar medium. In 36...

One of the most critical activities that PhD. Student maybe faced is selecting supervisor for her dissertation. Most often these decisions are in form a selection problem from finite number choices. The based on set criteria, such as professors' reputation, knowledge, and matching interests among others. However, application criteria done an unplanned manner, which can become one reasons regret, lack motivation, poor quality research output. need having who fits well with students'...

In this paper, we investigate the problem of system identification for autonomous Markov jump linear systems (MJS) with complete state observations. We propose switched least squares method MJS, show that is strongly consistent, and derive data-dependent data-independent rates convergence. particular, our rate convergence shows that, almost surely, error <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math...

Motivated by estimation problems arising in autonomous vehicles and decentralized control of unmanned aerial vehicles, we consider multi-agent filtering which multiple agents generate state estimates based on information the objective is to minimize a coupled mean-squared error call <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">team mean-square error</i> . We resulting as minimum team (MTMSE) estimates. show that MTMSE are different from...

The application of reinforcement learning to the optimal control building systems has gained traction in recent years as it can cut energy consumption and improve human comfort. Despite using sample-efficient algorithms, most related work requires several months sensor data operational parameters train an agent that outperforms existing rule-based controllers a large multi-zone building. Moreover, exploring state action spaces result poor indoor environmental quality for occupants. In this...

In this paper, we investigate the problem of system identification for autonomous Markov jump linear systems (MJS) with complete state observations. We propose switched least squares method MJS, show that is strongly consistent, and derive data-dependent data-independent rates convergence. particular, our rate convergence shows that, almost surely, error $\mathcal{O}(\sqrt {\log (T)/T} )$ where T time horizon. These results MJS has same as systems. compare those in literature present...

In this article, principal component analysis (PCA) is applied to improve Kalman state estimator performance in the presence of colored measurement noise without extending dimension. Unlike common methods proposed PCA-based doesn't use information dynamics. First, measurements Sensors are entered PCA block. The new data and previous ones, stored buffer, merged processed. output will be noiseless that increase accuracy estimator. An illustrative example simulated for comparisons standard...

The aim of this paper is to present mathematical models optimizing all materials flows in supply chain.In research a fuzzy multi-objective nonlinear mixed-integer programming model with piecewise linear membership function applied design multi echelon chain network (SCN) by considering total transportation costs and capacities echelons objectives.The that proposed study has 4 functions.The first minimizing the between (suppliers, factories, distribution centers (DCs) customers).The second...

One of the primary concerns on many countries is to determine different important factors affecting economic growth.In this paper, we study some such as unemployment rate, inflation ratio, population growth, average annual income, etc cluster countries.The proposed model paper uses analytical hierarchy process (AHP) prioritize criteria and then a K-mean technique 59 based ranked into four groups.The first group includes with high standards Germany Japan.In second cluster, there are...

A decentralized linear quadratic system with a major agent and collection of minor agents is considered. The affects the agents, but not vice versa. state observed by all agents. In addition, have noisy observation their local state. noise process <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">not</i> assumed to be Gaussian. structures optimal strategy best are characterized. It shown that agent's control action function minimum mean-squared...

We consider the problem of optimal decentralized estimation a linear stochastic process by multiple agents. Each agent receives noisy observation state and delayed observations its neighbors (according to pre-specified, strongly connected, communication graph). Based on their observations, all agents generate sequence estimates process. The objective is minimize total expected weighted mean square error between agents' over finite horizon. In centralized with criteria, estimator does not...

The aim of this research is to present a hybrid model select auto part suppliers.The proposed method paper uses factor analysis find the most influencing factors on maker selection and results are validated using different statistical tests such as Cronbach's Alpha Kaiser-Meyer.The analytical network process rank suppliers fuzzy goal programming choose appropriate alternative among various choices.The implementation used for case study real-world problem discussed.

Concurrent engineering (CE) is one of the widest known techniques for simultaneous planning product and process design.In concurrent engineering, design processes are often complicated with multiple conflicting criteria discrete sets feasible alternatives.Thus multi-criteria decision making (MCDM) integrated into CE to perform design.This paper proposes a framework governed by MCDM technique, which in conflict sense competing common resources achieve variously different performance...

We consider a static team problem in which agents observe correlated Gaussian observations and seek to minimize quadratic cost. It is assumed that the can be split into two parts: common are observed by all local individual agents. shown optimal strategies affine corresponding gains determined solving appropriate systems of linear equations. Two structures identified. The first may viewed as common-information based solution; second hierarchical control solution. A decentralized estimation...

This work studies the behaviors of two large-population teams competing in a discrete environment. The team-level interactions are modeled as zero-sum game while agent dynamics within each team is formulated collaborative mean-field problem. Drawing inspiration from literature, we first approximate with its infinite-population limit. Subsequently, construct fictitious centralized system and transform to an equivalent between coordinators. We study optimal coordination strategies for via...

In this paper, we consider learning and control problem in an unknown Markov jump linear system (MJLS) with perfect state observations. We first establish a generic upper bound on regret for any based algorithm. then propose certainty equivalence-based alagrithm show that algorithm achieves of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathcal{O}(\sqrt{T}{\mathrm{l}}\text{og}(T))$</tex> relative to certain subset the sample space. As part...