Finding the optimal dose combination in two-agent dose-finding trials is challenging due to limited sample sizes and the extensive range of potential doses.Unlike traditional chemotherapy or radiotherapy,which primari...Finding the optimal dose combination in two-agent dose-finding trials is challenging due to limited sample sizes and the extensive range of potential doses.Unlike traditional chemotherapy or radiotherapy,which primarily focuses on identifying the maximum tolerated dose(MTD),therapies involving targeted and immune agents facilitate the identifica-tion of an optimal biological dose combination(OBDC)by simultaneously evaluating both toxicity and efficacy.Cur-rently,most approaches to determining the OBDC in the literature are model-based and require complex model fittings,making them cumbersome and challenging to implement.To address these challenges,we developed a novel model-as-sisted approach called uTPI-Comb.This approach refines the established utility-based toxicity probability interval design by integrating a strategically devised zone-based local and global candidate set searching strategy,which can effectively optimize the decision-making process for two-agent dose escalation or de-escalation in drug combination trials.Extensive simulation studies demonstrate that the uTPI-Comb design speeds up the dose-searching process and provides substantial improvements over existing model-based methods in determining the optimal biological dose combinations.展开更多
NP-hard combinational optimization problem is not solved very well until now. One enhanced ants system based on ants system is advanced after analysis of the deficiencies of existing ants systems. Some improvements ar...NP-hard combinational optimization problem is not solved very well until now. One enhanced ants system based on ants system is advanced after analysis of the deficiencies of existing ants systems. Some improvements are made in state transfer rule and local modification rule. Furthermore, the enhanced ants system can solve NP-hard combinational optimization problem with restraints and condition path. The successful application of TSP problem and transportation net problem indicates that the proposed system has stronger function and higher efficiency than the original system.展开更多
The basic concepts and models of weapon-target assignment （WTA） are introduced and the mathematical nature of the WTA models is also analyzed. A systematic survey of research on WTA problem is provided. The present ...The basic concepts and models of weapon-target assignment （WTA） are introduced and the mathematical nature of the WTA models is also analyzed. A systematic survey of research on WTA problem is provided. The present research on WTA is focused on models and algorithms. In the research on models of WTA, the static WTA models are mainly studied and the dynamic WTA models are not fully studied in deed. In the research on algorithms of WTA, the intelligent algorithms are often used to solve the WTA problem. The small scale of static WTA problems has been solved very well, however, the large scale of dynamic WTA problems has not been solved effectively so far. Finally, the characteristics of dynamic WTA are analyzed and directions for the future research on dynamic WTA are discussed.展开更多
A novel algorithm for constructing support vector machine regression ensemble is proposed. As to regression prediction, support vector machine regression （SVMR） ensemble is proposed by resampling from given training...A novel algorithm for constructing support vector machine regression ensemble is proposed. As to regression prediction, support vector machine regression （SVMR） ensemble is proposed by resampling from given training data sets repeatedly and aggregating several independent SVMRs, each of which is trained to use a replicated training set. After training, several independently trained SVMRs need to be aggregated in an appropriate combination manner. Generally, the linear weighting is usually used like expert weighting score in Boosting Regression and it is without optimization capacity. Three combination techniques are proposed, including simple arithmetic mean, linear least square error weighting and nonlinear hierarchical combining that uses another upper-layer SVMR to combine several lower-layer SVMRs. Finally, simulation experiments demonstrate the accuracy and validity of the presented algorithm.展开更多
Reconnaissance mission planning of multiple unmanned aerial vehicles(UAVs)under an adversarial environment is a discrete combinatorial optimization problem which is proved to be a non-deterministic polynomial(NP)-comp...Reconnaissance mission planning of multiple unmanned aerial vehicles(UAVs)under an adversarial environment is a discrete combinatorial optimization problem which is proved to be a non-deterministic polynomial(NP)-complete problem.The purpose of this study is to research intelligent multiUAVs reconnaissance mission planning and online re-planning algorithm under various constraints in mission areas.For numerous targets scattered in the wide area,a reconnaissance mission planning and re-planning system is established,which includes five modules,including intelligence analysis,sub-mission area division,mission sequence planning,path smoothing,and online re-planning.The intelligence analysis module depicts the attribute of targets and the heterogeneous characteristic of UAVs and computes the number of sub-mission areas on consideration of voyage distance constraints.In the sub-mission area division module,an improved K-means clustering algorithm is designed to divide the reconnaissance mission area into several sub-mission areas,and each sub-mission is detected by the UAV loaded with various detective sensors.To control reconnaissance cost,the sampling and iteration algorithms are proposed in the mission sequence planning module,which are utilized to solve the optimal or approximately optimal reconnaissance sequence.In the path smoothing module,the Dubins curve is applied to smooth the flight path,which assure the availability of the planned path.Furthermore,an online re-planning algorithm is designed for the uncertain factor that the UAV is damaged.Finally,reconnaissance planning and re-planning experiment results show that the algorithm proposed in this paper are effective and the algorithms designed for sequence planning have a great advantage in solving efficiency and optimality.展开更多
基金This work was supported by the Natural Science Foundation of Anhui Province(2022AH050703)the National Natural Science Foundation of China(11671375).
文摘Finding the optimal dose combination in two-agent dose-finding trials is challenging due to limited sample sizes and the extensive range of potential doses.Unlike traditional chemotherapy or radiotherapy,which primarily focuses on identifying the maximum tolerated dose(MTD),therapies involving targeted and immune agents facilitate the identifica-tion of an optimal biological dose combination(OBDC)by simultaneously evaluating both toxicity and efficacy.Cur-rently,most approaches to determining the OBDC in the literature are model-based and require complex model fittings,making them cumbersome and challenging to implement.To address these challenges,we developed a novel model-as-sisted approach called uTPI-Comb.This approach refines the established utility-based toxicity probability interval design by integrating a strategically devised zone-based local and global candidate set searching strategy,which can effectively optimize the decision-making process for two-agent dose escalation or de-escalation in drug combination trials.Extensive simulation studies demonstrate that the uTPI-Comb design speeds up the dose-searching process and provides substantial improvements over existing model-based methods in determining the optimal biological dose combinations.
基金This project was supported by the National Natural Science Foundation of Shangdong Province (Y98F12093) .
文摘NP-hard combinational optimization problem is not solved very well until now. One enhanced ants system based on ants system is advanced after analysis of the deficiencies of existing ants systems. Some improvements are made in state transfer rule and local modification rule. Furthermore, the enhanced ants system can solve NP-hard combinational optimization problem with restraints and condition path. The successful application of TSP problem and transportation net problem indicates that the proposed system has stronger function and higher efficiency than the original system.
基金This project was supported by the National Defense Pre-Research Foundation of China
文摘The basic concepts and models of weapon-target assignment （WTA） are introduced and the mathematical nature of the WTA models is also analyzed. A systematic survey of research on WTA problem is provided. The present research on WTA is focused on models and algorithms. In the research on models of WTA, the static WTA models are mainly studied and the dynamic WTA models are not fully studied in deed. In the research on algorithms of WTA, the intelligent algorithms are often used to solve the WTA problem. The small scale of static WTA problems has been solved very well, however, the large scale of dynamic WTA problems has not been solved effectively so far. Finally, the characteristics of dynamic WTA are analyzed and directions for the future research on dynamic WTA are discussed.
基金The project was supported by the National Science Foundation of China (70572045)
文摘A novel algorithm for constructing support vector machine regression ensemble is proposed. As to regression prediction, support vector machine regression （SVMR） ensemble is proposed by resampling from given training data sets repeatedly and aggregating several independent SVMRs, each of which is trained to use a replicated training set. After training, several independently trained SVMRs need to be aggregated in an appropriate combination manner. Generally, the linear weighting is usually used like expert weighting score in Boosting Regression and it is without optimization capacity. Three combination techniques are proposed, including simple arithmetic mean, linear least square error weighting and nonlinear hierarchical combining that uses another upper-layer SVMR to combine several lower-layer SVMRs. Finally, simulation experiments demonstrate the accuracy and validity of the presented algorithm.
文摘Reconnaissance mission planning of multiple unmanned aerial vehicles(UAVs)under an adversarial environment is a discrete combinatorial optimization problem which is proved to be a non-deterministic polynomial(NP)-complete problem.The purpose of this study is to research intelligent multiUAVs reconnaissance mission planning and online re-planning algorithm under various constraints in mission areas.For numerous targets scattered in the wide area,a reconnaissance mission planning and re-planning system is established,which includes five modules,including intelligence analysis,sub-mission area division,mission sequence planning,path smoothing,and online re-planning.The intelligence analysis module depicts the attribute of targets and the heterogeneous characteristic of UAVs and computes the number of sub-mission areas on consideration of voyage distance constraints.In the sub-mission area division module,an improved K-means clustering algorithm is designed to divide the reconnaissance mission area into several sub-mission areas,and each sub-mission is detected by the UAV loaded with various detective sensors.To control reconnaissance cost,the sampling and iteration algorithms are proposed in the mission sequence planning module,which are utilized to solve the optimal or approximately optimal reconnaissance sequence.In the path smoothing module,the Dubins curve is applied to smooth the flight path,which assure the availability of the planned path.Furthermore,an online re-planning algorithm is designed for the uncertain factor that the UAV is damaged.Finally,reconnaissance planning and re-planning experiment results show that the algorithm proposed in this paper are effective and the algorithms designed for sequence planning have a great advantage in solving efficiency and optimality.
