This study explores a novel approach to path planning in deposition-based additive manufacturing, integrating the frequently overlooked process-induced temperature fields. Currently, existing ...approaches either ignore temperature effects entirely or only consider them in small-scale problems due to the high computational cost involved in predicting them and the combinatorial nature of path planning optimization. To address these challenges, the present work proposes an optimization pipeline that involves deriving a reduced order model from a finite volume method model with balanced truncation, using an analytical function to model the heat input and, calculating the steady-state response of the system to an arbitrary path using the Laplace transformation. Then, the optimization is transformed into a sequential decision-making problem and approximated with Monte Carlo tree search. The pipeline is validated through computational and experimental results, demonstrating its efficiency in managing large and complex geometries, as well as its resilience in overcoming the challenges posed by the simulation to reality gap.
Display omitted
•Path planning is transformed into a sequential decision-making problem.•Monte Carlo Tree Search (MCTS) is employed to approximate optimal paths.•The thermal dynamics of a part are approximated with a reduced order model (ROM).•The Laplace transformation is used to accelerate the evaluation of thermal fields.•The devised paths homogenize the temperature distributions of complex geometries.
Elastic Monte Carlo Tree Search Xu, Linjie; Dockhorn, Alexander; Perez-Liebana, Diego
IEEE transactions on games,
12/2023, Letnik:
15, Številka:
4
Journal Article
Recenzirano
Odprti dostop
Strategy games are a challenge for the design of AI agents due to their complexity and the combinatorial search space they produce. State abstraction has been applied in different domains to shrink ...the search space. Automatic state abstraction methods have gained much success in the planning domain and their transfer to strategy games raises a question of scalability. In this paper, we propose Elastic MCTS, an algorithm that uses automatic state abstraction to play strategy games. In Elastic MCTS, tree nodes are clustered dynamically. First, nodes are grouped by state abstraction for efficient exploration, to later be separated for refining exploitable action sequences. Such an elastic tree benefits from efficient information sharing while avoiding using an imperfect state abstraction during the whole search process. We provide empirical analyses of the proposed method in three strategy games of different complexity. Our empirical results show that in all games, Elastic MCTS outperforms MCTS baselines by a large margin, with a considerable search tree size reduction at the expense of small computation time. The code for reproducing the reported results can be found at https://github.com/GAIGResearch/Stratega .
In this paper we investigate an application of hybrid Monte Carlo Tree Search (MCTS) based algorithms to solving dynamic decision making problems.
We employ UCT (the most popular MCTS approach) in ...combination with well-known Resource Constrained Project Scheduling Problem (RCPSP) and Stochastic Resource Constrained Project Scheduling Problem (SRCPSP) solvers to devise strategies for a generic and highly dynamic version of RCPSP, which we call Risk-Aware Project Scheduling Problem (RAPSP). We compare these strategies’ performance with results of both pure MCTS approach and non-MCTS solvers for projects of varied characteristics. We reach a conclusion that proposed hybrid simulation-heuristic methods are promising approaches to dynamic decision making problems, RAPSP in particular. Consequently, we argue that more research effort should be directed to applications of MCTS algorithm outside the domain of game-playing, with which it is commonly associated.
At the same time, to the best of our knowledge, this paper is the first attempt at defining generalized SRCPSP model encompassing arbitrary risks and risk response / mitigation strategies as an optimization problem and applying Computational Intelligence methods to build fully-automated decision making systems. We strongly believe it to be a research direction worth further investigation, combining project scheduling, risk management and metaheuristic optimization techniques into a well-defined platform allowing direct comparisons of different strategies.
Display omitted
•A computer model explains ketogenesis from medium-chain triglycerides in humans.•Lipolysis in the upper GI tract had the strongest effect on the bioavailability.•Peak ketone ...concentrations are affected by gastric emptying.•Conversion rates of fatty acids to ketones are not rate-limiting in the addressed dose range.
At present, tricaprilin is used as a ketogenic source for the management of mild to moderate Alzheimer’s disease. After administration of the medium-chain triglyceride, tricaprilin is hydrolyzed to octanoic acid and further metabolized to ketones, acting as an alternative energy substrate for the brain. In this investigation, we developed a physiologically-based biopharmaceutics model simulating in vivo processes following the peroral administration of tricaprilin. The model includes multiple data sources to establish a partially verified framework for the simulation of plasma profiles. The input parameters were identified based on existing literature data and in vitro digestion studies. Model validation was conducted using the data from a phase I clinical trial. A partial parameter sensitivity analysis elucidated various influences on the plasma ketone levels that are mainly responsible for the therapeutic effects of tricaprilin. Based on our findings, we concluded that dispersibility and lipolysis of tricaprilin together with the gastric emptying patterns are limiting ketogenesis, while other steps such as the conversion of octanoic acid to ketone bodies play a minor role only.
A sensor management method for joint multitarget search and track problems is proposed, where a single user-defined parameter allows for a tradeoff between the two objectives. The multitarget density ...is propagated using the Poisson multi-Bernoulli mixture filter, which eliminates the need for a separate handling of undiscovered targets and provides the theoretical foundation for a unified search and track method. Monte Carlo simulations of two scenarios are used to evaluate the performance of the proposed method.
In this article, we address the problem of stochastic motion planning under partial observability, more specifically, how to navigate a mobile robot equipped with continuous range sensors, such as ...LIDAR. In contrast to many existing robotic motion planning methods, we explicitly consider the uncertainty of the robot state by modeling the system as a partially observable Markov decision process (POMDP). Recent work on general purpose POMDP solvers is typically limited to discrete observation spaces, and does not readily apply to the proposed problem due to the continuous measurements from LIDAR. In this article, we build upon an existing Monte Carlo tree search method, partially observable Monte Carlo planning (POMCP), and propose a new algorithm POMCP++. Our algorithm can handle continuous observation spaces with a novel measurement selection strategy. The POMCP++ algorithm overcomes overoptimism in the value estimation of a rollout policy by removing the implicit perfect state assumption at the rollout phase. We validate POMCP++ in theory by proving it is a Monte Carlo tree search algorithm. Through comparisons with other methods that can also be applied to the proposed problem, we show that POMCP++ yields significantly higher success rate and total reward.
The synthesis of monocarboxylate transporters (MCTs) can be stimulated by aerobic training, but few is known about this effect associated or not with non-voluntary daily activities. We examined the ...effect of eight weeks of aerobic training in MCTs on the skeletal muscle and hypothalamus of less or more physically active mice, which can be achieved by keeping them in two different housing models, a small cage (SC) and a large cage (LC).
Forty male C57BL/6J mice were divided into four groups. In each housing condition, mice were divided into untrained (N) and trained (T). For 8 weeks, the trained animals ran on a treadmill with an intensity equivalent to 80 % of the individual critical velocity (CV), considered aerobic capacity, 40 min/day, 5 times/week. Protein expression of MCTs was determined with fluorescence Western Blot.
T groups had higher hypothalamic MCT2 than N groups (ANOVA, P = 0.032). Significant correlations were detected between hypothalamic MCT2 and CV. There was a difference between the SC and LC groups in relation to MCT4 in the hypothalamus (LC > SC, P = 0.044). Trained mice housed in LC (but not SC-T) exhibited a reduction in MCT4 muscle (P < 0.001).
Our findings indicate that aerobically trained mice increased the expression of MCT2 protein in the hypothalamus, which has been related to the uptake of lactate in neurons. Changes in energy metabolism in physically active mice (kept in LC) may be related to upregulation of hypothalamic MCT4, probably participating in the regulation of satiety.
•Our study is new in demonstrating changes of hypothalamic MCTs in animals with different levels of muscle activity.•Aerobic training associated with a more active lifestyle may play an interesting role in affecting MCTs.•Only trained mice kept in a large cage (but not for trained mice housed in SC) exhibited high MCT4 in hypothalamus.•Aerobic training decreased MCT4 in soleus muscles of mice housed in large cage, but not mice housed in small cage.•Trained mice exhibited higher aerobic capacity and hypothalamic MCT2 protein expression than non-trained groups.
Triple negative breast cancers (TNBC) remain a major medical challenge due to poor prognosis and limited treatment options. Mesothelin is a glycosyl-phosphatidyl inositol-linked membrane protein with ...restricted normal expression and high level expression in a large proportion of TNBC, thus qualifying as an attractive target. Its overexpression in breast tumors has been recently correlated with a decreased disease-free survival and an increase of distant metastases. The objective of the study was to investigate the relevance of a bispecific antibody-based immunotherapy approach through mesothelin targeting and CD16 engagement using a Fab-like bispecific format (MesobsFab). Using two TNBC cell lines with different level of surface mesothelin and epithelial/mesenchymal phenotypes, we showed that,
, MesobsFab promotes the recruitment and penetration of NK cells into tumor spheroids, induces potent dose-dependent cell-mediated cytotoxicity of mesothelin-positive tumor cells, cytokine secretion, and decreases cell invasiveness. MesobsFab was able to induce cytotoxicity in resting human peripheral blood mononuclear cells (PBMC), mainly through its NK cells-mediated antibody dependent cell cytotoxicity (ADCC) activity.
, the anti-tumor effect of MesobsFab depends upon a threshold of MSLN density on target cells. Collectively our data support mesothelin as a relevant therapeutic target for the subset of TNBC that overexpresses mesothelin characterized by a low overall and disease-free survival as well as the potential of MesobsFab as antibody-based immunotherapeutics.
For decades the game playing algorithms of choice have been based on the mini-max algorithm and have had considerable success in many games, e.g., chess and checkers. Recently a new algorithmic ...paradigm called Monte-Carlo Tree Search (MCTS) has been discovered and has proven to perform well in games where mini-max has failed, most notably in the game of Go. Many view mini-max and MCTS based searches as competing and incompatible approaches. However, a hybrid technique using features of both mini-max and MCTS is possible. We call this algorithm MCTS-EPT (MCTS with early playout termination) and study it from the context of three different games: Amazons, Breakthrough, and Havannah. This paper expands and elaborates on work presented in 1 and 2.