The Heterogeneous Capacitated Vehicle Routing Problem(HCVRP),which involves efficiently routing vehicles with diverse capacities to fulfill various customer demands at minimal cost,poses an NP-hard challenge in combin...The Heterogeneous Capacitated Vehicle Routing Problem(HCVRP),which involves efficiently routing vehicles with diverse capacities to fulfill various customer demands at minimal cost,poses an NP-hard challenge in combinatorial optimization.Recently,reinforcement learning approaches such as 2D Array Pointer Networks(2D-Ptr)have demonstrated remarkable speed in decision-making by modeling multiple agents’concurrent choices as a sequence of consecutive actions.However,these learning-based models often struggle with generalization,meaning they cannot seamlessly adapt to new scenarios with varying numbers of vehicles or customers without retraining.Inspired by the potential of multi-teacher knowledge distillation to harness diverse knowledge from multiple sources and craft a comprehensive student model,we propose to enhance the generalization capability of 2D-Ptr through Multiple Teacher-forcing Knowledge Distillation(MTKD).We initially train 12 unique 2D-Ptr models under various settings to serve as teacher models.Subsequently,we randomly sample a teacher model and a batch of problem instances,focusing on those where the chosen teacher performed best.This teacher model then solves these instances,generating high-reward action sequences to guide knowledge transfer to the student model.We conduct rigorous evaluations across four distinct datasets,each comprising four HCVRP instances of varying scales.Our empirical findings underscore the proposed method superiority over existing learning-based methods in terms of both computational efficiency and solution quality.展开更多
Micro-nano Earth Observation Satellite(MEOS)constellation has the advantages of low construction cost,short revisit cycle,and high functional density,which is considered a promising solution for serving rapidly growin...Micro-nano Earth Observation Satellite(MEOS)constellation has the advantages of low construction cost,short revisit cycle,and high functional density,which is considered a promising solution for serving rapidly growing observation demands.The observation Scheduling Problem in the MEOS constellation(MEOSSP)is a challenging issue due to the large number of satellites and tasks,as well as complex observation constraints.To address the large-scale and complicated MEOSSP,we develop a Two-Stage Scheduling Algorithm based on the Pointer Network with Attention mechanism(TSSA-PNA).In TSSA-PNA,the MEOS observation scheduling is decomposed into a task allocation stage and a single-MEOS scheduling stage.In the task allocation stage,an adaptive task allocation algorithm with four problem-specific allocation operators is proposed to reallocate the unscheduled tasks to new MEOSs.Regarding the single-MEOS scheduling stage,we design a pointer network based on the encoder-decoder architecture to learn the optimal singleMEOS scheduling solution and introduce the attention mechanism into the encoder to improve the learning efficiency.The Pointer Network with Attention mechanism(PNA)can generate the single-MEOS scheduling solution quickly in an end-to-end manner.These two decomposed stages are performed iteratively to search for the solution with high profit.A greedy local search algorithm is developed to improve the profits further.The performance of the PNA and TSSA-PNA on singleMEOS and multi-MEOS scheduling problems are evaluated in the experiments.The experimental results demonstrate that PNA can obtain the approximate solution for the single-MEOS scheduling problem in a short time.Besides,the TSSA-PNA can achieve higher observation profits than the existing scheduling algorithms within the acceptable computational time for the large-scale MEOS scheduling problem.展开更多
The existing abstractive text summarisation models only consider the word sequence correlations between the source document and the reference summary,and the summary generated by models lacks the cover of the subject ...The existing abstractive text summarisation models only consider the word sequence correlations between the source document and the reference summary,and the summary generated by models lacks the cover of the subject of source document due to models'small perspective.In order to make up these disadvantages,a multi‐domain attention pointer(MDA‐Pointer)abstractive summarisation model is proposed in this work.First,the model uses bidirectional long short‐term memory to encode,respectively,the word and sentence sequence of source document for obtaining the semantic representations at word and sentence level.Furthermore,the multi‐domain attention mechanism between the semantic representations and the summary word is established,and the proposed model can generate summary words under the proposed attention mechanism based on the words and sen-tences.Then,the words are extracted from the vocabulary or the original word sequences through the pointer network to form the summary,and the coverage mechanism is introduced,respectively,into word and sentence level to reduce the redundancy of sum-mary content.Finally,experiment validation is conducted on CNN/Daily Mail dataset.ROUGE evaluation indexes of the model without and with the coverage mechanism are improved respectively,and the results verify the validation of model proposed by this paper.展开更多
Few-shot named entity recognition(NER)aims to identify named entities in new domains using a limited amount of annotated data.Previous methods divided this task into entity span detection and entity classification,ach...Few-shot named entity recognition(NER)aims to identify named entities in new domains using a limited amount of annotated data.Previous methods divided this task into entity span detection and entity classification,achieving good results.However these methods are limited by the imbalance between the entity and non-entity categories due to the use of sequence labeling for entity span detection.To this end,a point-proto network(PPN)combining pointer and prototypical networks was proposed.Specifically,the pointer network generates the position of entities in sentences in the entity span detection stage.The prototypical network builds semantic prototypes of entity types and classifies entities based on their distance from these prototypes in the entity classification stage.Moreover,the low-rank adaptation(LoRA)fine-tuning method,which involves freezing the pre-trained weights and injecting a trainable decomposition matrix,reduces the parameters that need to be trained and saved.Extensive experiments on the few-shot NER Dataset(Few-NERD)and Cross-Dataset demonstrate the superiority of PPN in this domain.展开更多
Shuttle tankers scheduling is an important task in offshore oil and gas transportation process,which involves operating time window fulfillment,optimal transportation planning,and proper inventory management.However,c...Shuttle tankers scheduling is an important task in offshore oil and gas transportation process,which involves operating time window fulfillment,optimal transportation planning,and proper inventory management.However,conventional approaches like Mixed lnteger Linear Programming(MlLP)or meta heuristic algorithms often fail in long running time.In this paper,a Graph Pointer Network(GPN)based Hierarchical Curriculum Reinforcement Learning(HCRl)method is proposed to solve Shuttle Tankers Scheduling Problem(STSP)The model is trained to divide STSP into voyage and operation stages and generate routing and inventory management decisions sequentially.An asynchronous training strategy is developed to address the coupling between stages.Comparison experiments demonstrate that the proposed HCRL method achieves 12%shortel tour lengths on average compared to heuristic algorithms.Additional experiments validate its generalizability to unseen instances and scalability to larger instances.展开更多
Combinatorial Optimization Problems(COPs)are a class of optimization problems that are commonly encountered in industrial production and everyday life.Over the last few decades,traditional algorithms,such as exact alg...Combinatorial Optimization Problems(COPs)are a class of optimization problems that are commonly encountered in industrial production and everyday life.Over the last few decades,traditional algorithms,such as exact algorithms,approximate algorithms,and heuristic algorithms,have been proposed to solve COPs.However,as COPs in the real world become more complex,traditional algorithms struggle to generate optimal solutions in a limited amount of time.Since Deep Neural Networks(DNNs)are not heavily dependent on expert knowledge and are adequately flexible for generalization to various COPs,several DNN-based algorithms have been proposed in the last ten years for solving COPs.Herein,we categorize these algorithms into four classes and provide a brief overview of their applications in real-world problems.展开更多
基金in part by the National Science Foundation of China under Grant No.62276238in part by the National Science Foundation for Distinguished Young Scholars of China under Grant No.62325602in part by the Natural Science Foundation of Henan,China under Grant No.232300421095.
文摘The Heterogeneous Capacitated Vehicle Routing Problem(HCVRP),which involves efficiently routing vehicles with diverse capacities to fulfill various customer demands at minimal cost,poses an NP-hard challenge in combinatorial optimization.Recently,reinforcement learning approaches such as 2D Array Pointer Networks(2D-Ptr)have demonstrated remarkable speed in decision-making by modeling multiple agents’concurrent choices as a sequence of consecutive actions.However,these learning-based models often struggle with generalization,meaning they cannot seamlessly adapt to new scenarios with varying numbers of vehicles or customers without retraining.Inspired by the potential of multi-teacher knowledge distillation to harness diverse knowledge from multiple sources and craft a comprehensive student model,we propose to enhance the generalization capability of 2D-Ptr through Multiple Teacher-forcing Knowledge Distillation(MTKD).We initially train 12 unique 2D-Ptr models under various settings to serve as teacher models.Subsequently,we randomly sample a teacher model and a batch of problem instances,focusing on those where the chosen teacher performed best.This teacher model then solves these instances,generating high-reward action sequences to guide knowledge transfer to the student model.We conduct rigorous evaluations across four distinct datasets,each comprising four HCVRP instances of varying scales.Our empirical findings underscore the proposed method superiority over existing learning-based methods in terms of both computational efficiency and solution quality.
基金supported by the National Natural Science Foundation of China(No.62101587)the National Funded Postdoctoral Researcher Program of China(No.GZC20233578)。
文摘Micro-nano Earth Observation Satellite(MEOS)constellation has the advantages of low construction cost,short revisit cycle,and high functional density,which is considered a promising solution for serving rapidly growing observation demands.The observation Scheduling Problem in the MEOS constellation(MEOSSP)is a challenging issue due to the large number of satellites and tasks,as well as complex observation constraints.To address the large-scale and complicated MEOSSP,we develop a Two-Stage Scheduling Algorithm based on the Pointer Network with Attention mechanism(TSSA-PNA).In TSSA-PNA,the MEOS observation scheduling is decomposed into a task allocation stage and a single-MEOS scheduling stage.In the task allocation stage,an adaptive task allocation algorithm with four problem-specific allocation operators is proposed to reallocate the unscheduled tasks to new MEOSs.Regarding the single-MEOS scheduling stage,we design a pointer network based on the encoder-decoder architecture to learn the optimal singleMEOS scheduling solution and introduce the attention mechanism into the encoder to improve the learning efficiency.The Pointer Network with Attention mechanism(PNA)can generate the single-MEOS scheduling solution quickly in an end-to-end manner.These two decomposed stages are performed iteratively to search for the solution with high profit.A greedy local search algorithm is developed to improve the profits further.The performance of the PNA and TSSA-PNA on singleMEOS and multi-MEOS scheduling problems are evaluated in the experiments.The experimental results demonstrate that PNA can obtain the approximate solution for the single-MEOS scheduling problem in a short time.Besides,the TSSA-PNA can achieve higher observation profits than the existing scheduling algorithms within the acceptable computational time for the large-scale MEOS scheduling problem.
基金supported by the National Social Science Foundation of China(2017CG29)the Science and Technology Research Project of Chongqing Municipal Education Commission(2019CJ50)the Natural Science Foundation of Chongqing(2017CC29).
文摘The existing abstractive text summarisation models only consider the word sequence correlations between the source document and the reference summary,and the summary generated by models lacks the cover of the subject of source document due to models'small perspective.In order to make up these disadvantages,a multi‐domain attention pointer(MDA‐Pointer)abstractive summarisation model is proposed in this work.First,the model uses bidirectional long short‐term memory to encode,respectively,the word and sentence sequence of source document for obtaining the semantic representations at word and sentence level.Furthermore,the multi‐domain attention mechanism between the semantic representations and the summary word is established,and the proposed model can generate summary words under the proposed attention mechanism based on the words and sen-tences.Then,the words are extracted from the vocabulary or the original word sequences through the pointer network to form the summary,and the coverage mechanism is introduced,respectively,into word and sentence level to reduce the redundancy of sum-mary content.Finally,experiment validation is conducted on CNN/Daily Mail dataset.ROUGE evaluation indexes of the model without and with the coverage mechanism are improved respectively,and the results verify the validation of model proposed by this paper.
基金supported by the National Key Research and Development Project(2021YFF0901701)。
文摘Few-shot named entity recognition(NER)aims to identify named entities in new domains using a limited amount of annotated data.Previous methods divided this task into entity span detection and entity classification,achieving good results.However these methods are limited by the imbalance between the entity and non-entity categories due to the use of sequence labeling for entity span detection.To this end,a point-proto network(PPN)combining pointer and prototypical networks was proposed.Specifically,the pointer network generates the position of entities in sentences in the entity span detection stage.The prototypical network builds semantic prototypes of entity types and classifies entities based on their distance from these prototypes in the entity classification stage.Moreover,the low-rank adaptation(LoRA)fine-tuning method,which involves freezing the pre-trained weights and injecting a trainable decomposition matrix,reduces the parameters that need to be trained and saved.Extensive experiments on the few-shot NER Dataset(Few-NERD)and Cross-Dataset demonstrate the superiority of PPN in this domain.
基金supported by the National Natural Science Foundation of China(Nos.22178383 and 21706282)Beijing Natural Science Foundation(No.2232021)Research Foundation of China University of Petroleum(Beijing)(No.2462020BJRC004).
文摘Shuttle tankers scheduling is an important task in offshore oil and gas transportation process,which involves operating time window fulfillment,optimal transportation planning,and proper inventory management.However,conventional approaches like Mixed lnteger Linear Programming(MlLP)or meta heuristic algorithms often fail in long running time.In this paper,a Graph Pointer Network(GPN)based Hierarchical Curriculum Reinforcement Learning(HCRl)method is proposed to solve Shuttle Tankers Scheduling Problem(STSP)The model is trained to divide STSP into voyage and operation stages and generate routing and inventory management decisions sequentially.An asynchronous training strategy is developed to address the coupling between stages.Comparison experiments demonstrate that the proposed HCRL method achieves 12%shortel tour lengths on average compared to heuristic algorithms.Additional experiments validate its generalizability to unseen instances and scalability to larger instances.
基金supported by the National Natural Science Foundation of China(Nos.62173258 and 61773296).
文摘Combinatorial Optimization Problems(COPs)are a class of optimization problems that are commonly encountered in industrial production and everyday life.Over the last few decades,traditional algorithms,such as exact algorithms,approximate algorithms,and heuristic algorithms,have been proposed to solve COPs.However,as COPs in the real world become more complex,traditional algorithms struggle to generate optimal solutions in a limited amount of time.Since Deep Neural Networks(DNNs)are not heavily dependent on expert knowledge and are adequately flexible for generalization to various COPs,several DNN-based algorithms have been proposed in the last ten years for solving COPs.Herein,we categorize these algorithms into four classes and provide a brief overview of their applications in real-world problems.
