Article
Computer Science, Artificial Intelligence
Channel A. Rodriguez, Phillip R. Jenkins, Matthew J. Robbins
Summary: This paper focuses on the MEDEVAC dispatching problem in combat operations, considering triage classification errors and the possibility of having blood transfusion kits on board select MEDEVAC units. A Markov decision process model is formulated and approximate dynamic programming techniques are used to develop high-quality policies. Results show that applying this technique can improve life-saving performance by up to 29%. This research is important for the military medical community and can guide future military MEDEVAC operations.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Automation & Control Systems
Alberto Maria Metelli, Matteo Pirotta, Daniele Calandriello, Marcello Restelli
Summary: This paper presents a study on the policy improvement step in approximate policy iteration algorithms, proposing three safe policy-iteration schemas to address oscillations in policy iteration. The proposed algorithms are empirically evaluated and compared in various domains to explore solutions for potential issues in policy iteration.
JOURNAL OF MACHINE LEARNING RESEARCH
(2021)
Article
Thermodynamics
Qirun Sun, Zhi Wu, Wei Gu, Tao Zhu, Lei Zhong, Ting Gao
Summary: This study proposes a flexible distribution system expansion planning model based on approximate dynamic programming, taking into account long-term load growth uncertainty and short-term power fluctuations, and developing a flexible investment strategy using Markov decision process. Case studies show the feasibility and benefits of the proposed planning approach in significantly reducing investment risks and configuring renewable energy equipment more reasonably.
Article
Automation & Control Systems
Xinyan Ou, Qing Chang, Nilanjan Chakraborty
Summary: This article proposes an innovative method, Q-ADP, that integrates reinforcement learning and approximate dynamic programming for real-time gantry scheduling in a gantry work cell. Numerical studies show that Q-ADP outperforms standard Q-learning and requires less data for convergence. By learning directly from interactions with the environment, the method avoids bias from model designing, making it particularly useful when real data are limited.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2021)
Article
Thermodynamics
Houwang Zhang, Qiuwei Wu, Jian Chen, Lina Lu, Jiangfeng Zhang, Shuyi Zhang
Summary: To achieve the coordinated development of electricity and gas systems, a distributed approximate dynamic programming based multiple stage stochastic planning scheme is proposed in this paper. The scheme combines the approximate dynamic programming method and the alternating direction multiplier method to coordinate the planning of power grid, natural gas pipelines, substations, photovoltaics, wind turbines, gas turbines, and energy storage devices. The scheme considers both long-term and short-term uncertainties, and is validated through numerical tests.
Article
Automation & Control Systems
Jayakumar Subramanian, Amit Sinha, Raihan Seraj, Aditya Mahajan
Summary: This paper proposes a theoretical framework for approximate planning and learning in partially observed systems, based on the concept of information state. By defining two types of information state, the authors demonstrate their applications and properties, and prove the approximate optimality of policies computed using approximate information states. Additionally, the paper explores several approximations in state, observation, and action spaces, and presents an AIS-based policy gradient algorithm.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Engineering, Industrial
Arturo Wenzel, Antoine Saure, Alejandro Cataldo, Pablo A. Rey, Cesar Sanchez
Summary: A solution approach is proposed to increase the cost-efficiency of system-wide capacity use in chemotherapy sessions assignment. The approach allows patients to be treated at centers other than their home center, resulting in a 20% reduction in operating costs and a halving of existing first-session waiting times. However, it is important to implement the proposed proactive assignment policy for the network-based scheduling procedure to bring real benefits.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2023)
Article
Management
Faisal Alkaabneh, Ali Diabat, Huaizhu Oliver Gao
Summary: This article proposes a framework for optimizing the allocation of resources by food banks to address the impact of food insecurity and poor nutrition on health issues. Through a dynamic programming model and computational experiments, it is shown that this approach significantly improves total utility and the nutrition of the served population.
OMEGA-INTERNATIONAL JOURNAL OF MANAGEMENT SCIENCE
(2021)
Article
Computer Science, Interdisciplinary Applications
Jian Zhang, Mahjoub Dridi, Abdellah El Moudni
Summary: The paper proposes an approximate dynamic programming algorithm to solve the admission control problem for elective patients, achieving a balance between waiting times and resource utilization. The algorithm efficiently manages a waiting list with dynamic priority scores, providing high-quality near-optimal policies for realistically sized problems.
COMPUTERS & OPERATIONS RESEARCH
(2021)
Article
Transportation Science & Technology
Hoa T. M. Nguyen, Andy H. F. Chow
Summary: This paper presents an adaptive optimization framework for dynamic rail transit network operations using a rollout surrogate-approximate dynamic programming method. The proposed framework reduces passengers' waiting times significantly with reasonable computational time. The results suggest the potential of the proposed optimizer for real-time applications in large-scale rail transit networks.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2023)
Article
Engineering, Industrial
Ipek Kivanc, Demet Ozgur-Unluakin, Taner Bilgic
Summary: This study surveyed POMDP solution approaches and solvers, compared them using experimental models with different complexities, and modeled the maintenance problem of a one-line regenerative air heater system using factored POMDPs. Sensitivity analyses were performed on the obtained policy, showing the advantages of factored POMDPs in multi-component system maintenance.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Computer Science, Interdisciplinary Applications
Phillip R. Jenkins, Matthew J. Robbins, Brian J. Lunday
Summary: Military medical planners must consider efficient dispatching of aerial medical evacuation assets to support high-intensity combat operations. A Markov decision process model is formulated to address this issue, and approximate dynamic programming methods are used to find high-quality dispatching policies. Results show that the NN-API policies significantly outperform existing benchmark policies in most problem instances.
INFORMS JOURNAL ON COMPUTING
(2021)
Article
Operations Research & Management Science
Aaron Sidford, Mengdi Wang, Xian Wu, Yinyu Ye
Summary: This paper provides faster algorithms for approximately solving discounted Markov decision processes in multiple parameter regimes. The algorithms achieve linear time and linear convergence, improving upon previous best algorithms. By cleverly modifying approximate value iteration and combining classic analysis with variance reduction techniques, the paper ensures monotonic progress towards the optimal value and utilizes sampling to obtain linearly convergent linear programming algorithms.
NAVAL RESEARCH LOGISTICS
(2023)
Article
Management
Phillip R. Jenkins, Matthew J. Robbins, Brian J. Lunday
Summary: This research aims to optimize the performance of the US Army MEDEVAC systems by analyzing and solving the MEDEVAC DPR problem.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2021)
Article
Engineering, Multidisciplinary
Fang Xie, Haitao Li, Zhe Xu
Summary: The study focuses on the stochastic resource-constrained project scheduling problem with uncertain resource availability, proposing a new MDP model and an ADP algorithm to handle insufficient resource capacity, with proved theoretical sequential improvement property.
APPLIED MATHEMATICAL MODELLING
(2021)
Article
Computer Science, Theory & Methods
Nahla Ben Amor, Zeineb El Khalfi, Helene Fargier, Regis Sabbadin
FUZZY SETS AND SYSTEMS
(2019)
Article
Multidisciplinary Sciences
Hui Xiao, Eve McDonald-Madden, Regis Sabbadin, Nathalie Peyrard, Laura E. Dee, Ladine Chades
NATURE COMMUNICATIONS
(2019)
Article
Forestry
Isak Lodin, Ljusk Ola Eriksson, Nicklas Forsell, Anu Korosuo
Article
Forestry
Ljusk Ola Eriksson, Nicklas Forsell, Jeannette Eggers, Tord Snall
Article
Forestry
Mykola Gusti, Fulvio Di Fulvio, Peter Biber, Anu Korosuo, Nicklas Forsell
Article
Ecology
Peter Biber, Adam Felton, Maarten Nieuwenhuis, Matts Lindbladh, Kevin Black, Jan Bahyl, Ozkan Bingol, Jose G. Borges, Brigite Botequim, Vilis Brukas, Miguel N. Bugalho, Giulia Corradini, Ljusk Ola Eriksson, Nicklas Forsell, Geerten M. Hengeveld, Marjanke A. Hoogstra-Klein, Ali Ihsan Kadiogullari, Uzay Karahalil, Isak Lodin, Anders Lundholm, Ekaterina Makrickiene, Mauro Masiero, Gintautas Mozgeris, Nerijus Pivoriunas, Werner Poschenrieder, Hans Pretzsch, Robert Sedmak, Jan Tucek
FRONTIERS IN ECOLOGY AND EVOLUTION
(2020)
Article
Environmental Sciences
Stefan Frank, Mykola Gusti, Petr Havlik, Pekka Lauri, Fulvio DiFulvio, Nicklas Forsell, Tomoko Hasegawa, Tamas Krisztin, Amanda Palazzo, Hugo Valin
Summary: The article assesses the implications of achieving selected key SDG indicators on land-based climate change mitigation potential, highlighting the impact of protecting highly biodiverse ecosystems on biomass potentials and the synergies with greenhouse gas abatement achieved through SDGs. The study suggests that achieving SDGs could help realize up to 25% of the expected greenhouse gas abatement from land use, necessary to stay on track with the 1.5 degrees C target until 2050, without additional mitigation policies. Future land use mitigation policies should consider and take advantage of these synergies across SDGs.
ENVIRONMENTAL RESEARCH LETTERS
(2021)
Article
Environmental Sciences
Moonil Kim, Florian Kraxner, Nicklas Forsell, Cholho Song, Woo-Kyun Lee
Summary: This study examines the integrated impact of adaptive management and climate change on forest ecosystem services in South Korea using spatially explicit tools. Results indicate trade-offs between industrial wood production and freshwater supply, and between forest recreation and forest carbon storage. Harvest activity may have short-term negative effects on carbon sequestration, but could be positive in the long term through reforestation activities. Future climate change in Korea until 2050 is projected to have a generally negative influence on forest carbon sequestration, which could be partially offset through harvest management activities to maintain equilibrium in ecosystem services.
REGIONAL ENVIRONMENTAL CHANGE
(2021)
Review
Green & Sustainable Science & Technology
Hanna Fekete, Takeshi Kuramochi, Mark Roelfsema, Michel den Elzen, Nicklas Forsell, Niklas Hoehne, Lisa Luna, Frederic Hans, Sebastian Sterl, Jos Olivier, Heleen van Soest, Stefan Frank, Mykola Gusti
Summary: This article reviews climate change mitigation policies in China, the European Union, India, Japan, and the United States, focusing on their historical performance and target goals in various sectors. While most countries have successful policies in renewable energy, fuel efficiency, electrification of passenger vehicles, and forestry, there are still areas with limited information or comprehensive policies, such as buildings and agriculture. The study suggests that transformative policies are needed to achieve global emissions reductions in line with the goals of the Paris Agreement.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2021)
Article
Environmental Sciences
Takeshi Kuramochi, Leonardo Nascimento, Mia Moisio, Michel den Elzen, Nicklas Forsell, Heleen van Soest, Paola Tanguy, Sofia Gonzales, Frederic Hans, M. Louise Jeffery, Hanna Fekete, Tessa Schiefer, Maria Jose de Villafranca Casas, Gustavo De Vivero-Serrano, Ioannis Dafnomilis, Mark Roelfsema, Niklas Hoehne
Summary: This study compares greenhouse gas emissions projections for nine key non-G20 countries in 2030, finding that some countries may meet or exceed their targets, while others will need to strengthen their actions to achieve their goals.
ENVIRONMENTAL SCIENCE & POLICY
(2021)
Correction
Environmental Sciences
Takeshi Kuramochi, Leonardo Nascimento, Mia Moisio, Michel den Elzen, Nicklas Forsell, Heleen van Soest, Paola Tanguy, Sofia Gonzales, Frederic Hans, M. Louise Jeffery, Hanna Fekete, Tessa Schiefer, Maria Jose de Villafranca Casas, Gustavo De Vivero-Serrano, Ioannis Dafnomilis, Mark Roelfsema, Niklas Hohne
ENVIRONMENTAL SCIENCE & POLICY
(2021)
Article
Economics
Pekka Lauri, Nicklas Forsell, Fulvio Di Fulvio, Tord Snall, Petr Havlik
Summary: This study investigates the impact of material substitution between C, NC and R biomass on forest industry raw material use and regional competitiveness. It shows that an increase in the availability of R biomass would allow traditional forest industry regions to maintain their competitiveness, while a perfect substitution between C and NC biomass would decrease their competitiveness and increase that of emerging forest industry regions such as South America, Asia and Africa.
FOREST POLICY AND ECONOMICS
(2021)
Article
Ecology
Marie-Josee Cros, Jean-Noel Aubertot, Sabrina Gaba, Xavier Reboud, Regis Sabbadin, Nathalie Peyrard
Summary: Conventional pest management relies on pesticides, but their negative externalities are known. Sustainable practices like Integrated Pest Management are essential to limit crop damage. Pest monitoring networks provide crucial information, with effectiveness depending on spatial resolution and memory length. Optimizing PMNs is complex, as seen in a theoretical model's comparison of different PMNs' performances. Increasing spatial resolution can reduce treatments for endocyclic pests, while past observations and PMN information have less impact on non-endocyclic pests.
THEORETICAL POPULATION BIOLOGY
(2021)
Review
Agriculture, Multidisciplinary
Romain Gautron, Odalric-Ambrym Maillard, Philippe Preux, Marc Corbeels, Regis Sabbadin
Summary: Reinforcement learning is a branch of machine learning that deals with sequential decision-making in uncertain environments. It has the potential to address some of the criticisms of crop management decision support systems, but its application in this field is currently limited. Further research and collaboration between the reinforcement learning and agronomy communities are needed to fully explore its potential in agricultural decision-making.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Ecology
Sam Nicol, Marie-Josee Cros, Nathalie Peyrard, Regis Sabbadin, Ronan Trepos, Richard A. Fuller, Bradley K. Woodworth
Summary: This article introduces the concept of FlywayNet, a discrete network model based on observed count data, to determine the structure of migratory networks in birds. By modeling noisy observations and flexible stopover durations using interacting hidden semi-Markov models, this approach advances previous studies and provides a flexible framework for studying migratory networks in birds and other organisms.
METHODS IN ECOLOGY AND EVOLUTION
(2023)
Article
Computer Science, Artificial Intelligence
Timotheus Kampik, Kristijonas Cyras, Jose Ruiz Alarcon
Summary: This paper presents a formal approach to explaining changes in inference in Quantitative Bipolar Argumentation Frameworks (QBAFs). The approach traces the causes of strength inconsistencies and provides explanations for them.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Xiangnan Zhou, Longchun Wang, Qingguo Li
Summary: This paper aims to establish a closer connection between domain theory and Formal Concept Analysis (FCA) by introducing the concept of an optimized concept for a formal context. With the utilization of optimized concepts, it is demonstrated that the class of formal contexts directly corresponds to algebraic domains. Additionally, two subclasses of formal contexts are identified to characterize algebraic L-domains and Scott domains. An application is presented to address the open problem of reconstructing bounded complete continuous domains using attribute continuous contexts, and the presentation of algebraic domains is extended to a categorical equivalence.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Sihan Wang, Zhong Yuan, Chuan Luo, Hongmei Chen, Dezhong Peng
Summary: Anomaly detection is widely used in various fields, but most current methods only work for specific data and ignore uncertain information such as fuzziness. This paper proposes an anomaly detection algorithm based on fuzzy rough entropy, which effectively addresses the similarity between high-dimensional objects using distance and correlation measures. The algorithm is compared and analyzed with mainstream anomaly detection algorithms on publicly available datasets, showing superior performance and flexibility.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Mario Alviano, Francesco Bartoli, Marco Botta, Roberto Esposito, Laura Giordano, Daniele Theseider Dupre
Summary: This paper investigates the relationships between a multipreferential semantics in defeasible reasoning and a multilayer neural network model. Weighted knowledge bases are considered for a simple description logic with typicality under a concept-wise multipreference semantics. The semantics is used to interpret MultiLayer Perceptrons (MLPs) preferentially. Model checking and entailment based approach are employed in verifying conditional properties of MLPs.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Bazin Alexandre, Galasso Jessie, Kahn Giacomo
Summary: Formal concept analysis is a mathematical framework that represents the information in binary object-attribute datasets using a lattice of formal concepts. It has been extended to handle more complex data types, such as relational data and n-ary relations. This paper presents a framework for polyadic relational concept analysis, which extends relational concept analysis to handle relational datasets consisting of n-ary relations.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Ander Gray, Marcelo Forets, Christian Schilling, Scott Ferson, Luis Benet
Summary: The presented method combines reachability analysis and probability bounds analysis to handle imprecisely known random variables. It can rigorously compute the temporal evolution of p-boxes and provide interval probabilities for formal verification problems. The method does not impose strict constraints on the input probability distribution or p-box and can handle multivariate p-boxes with a consonant approximation method.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)
Article
Computer Science, Artificial Intelligence
Laszlo Csato
Summary: This paper studies a special type of incomplete pairwise comparison matrices and proposes a new method to determine the missing elements without violating the ordinal property.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2024)