Birdie Blog

Thinking will not overcome fear but action will.

Towards Generalizable Neural Solvers for Vehicle Routing Problems via Ensemble with Transferrable Local Policy

IJCAI2024 局部策略 泛化能力

Towards Generalizable Neural Solvers for Vehicle Routing Problems via Ensemble with Transferrable Local Policy 南京大学,华为诺亚方舟实验室 IJCAI 2024 代码:https://github.com/lamda-bbo/ELG 摘要 机器学习已经被用来帮助解决NP-...

Cross-Problem Learning for Solving Vehicle Routing Problems

IJCAI2024 模型微调 迁移学习

Cross-Problem Learning for Solving Vehicle Routing Problems 求解车辆路线问题的交叉问题学习 来自:A*STAR IJCAI 2024 代码:https://github.com/Zhuoyi-Lin/Cross_problem_learning. 摘要 现有的神经启发式算法通常针对每个特定的车辆路径问题(VRP)从头开始...

The Faiss Library

一个用于向量检索的C++库

The Faiss Library FAIR, Meta 2024 摘要 向量数据库管理嵌入向量的大型集合。随着人工智能应用的快速增长,需要存储和索引的嵌入数量也在快速增长。Faiss库致力于向量相似性搜索,这是向量数据库的核心功能。Faiss是一个索引方法和相关原语的工具包,用于搜索、聚类、压缩和转换向量。本文首先介绍了矢量搜索的权衡空间,然后从结构、优化方法和接口等方面介绍了Fa...

Deep reinforcement learning for the dynamic and uncertain vehicle routing problem

动态不确定VRP

Deep reinforcement learning for the dynamic and uncertain vehicle routing problem 2023 Applied Intelligence 福州大学经济与管理学院 摘要 对现实城市物流进行准确、实时的跟踪已成为智能交通领域的热门研究课题。而城市物流服务的路径选择通常是通过复杂的数学和分析方法来完成的。然而,现...

TAP:Transparent and Privacy-Preserving Data Services

隐私保护 透明日志

TAP: Transparent and Privacy-Preserving Data Services 出自:32nd USENIX Security Symposium (USENIX Security 23) 摘要 如今的用户期望从处理他们数据的服务中获得更多的安全性。除了传统的数据隐私和完整性要求外,他们还期望透明度,即服务对数据的处理可以由用户和受信任的审计员进行验证。...

Merkle2:A Low-Latency Transparency Log System

隐私保护 透明日志

Merkle2:A Low-Latency Transparency Log System 2021 IEEE Symposium on Security and Privacy (SP) Yuncong Hu, Kian Hooshmand, Harika Kalidhindi, Seung Jin Yang*, Raluca Ada Popa 加利福利亚大学,伯克利 摘要 透明...

H-TSP Hierarchically Solving the Large-Scale Travelling Salesman Problem

AAAI2023 分治 大规模TSP

H-TSP Hierarchically Solving the Large-Scale Travelling Salesman Problem AAAI 2023 杭州电子科技大学和微软亚洲研究所 摘要 本文提出了一个基于分层强化学习的端到端学习框架,称为H-TSP,用于解决大规模旅行商问题(TSP)。该方法包含两个策略:上层策略用于将原始的规模较大的问题拆解成多个规模较小的子问题...

Compositional Messagepassing Neural Network (CMPNN)

IJCAI2020 CMPNN

Compositional Messagepassing Neural Network (CMPNN) 选自: Communicative Representation Learning on Attributed Molecular Graphs(IJCAI20) A reinforcement learning approach for optimizing multipl...

SplitNet A Reinforcement Learning Based Sequence Splitting Method for the MinMax Multiple Travelling Salesman Problem

AAAI2023 从TSP重构MinMax mTSP

SplitNet: A Reinforcement Learning Based Sequence Splitting Method for the MinMax Multiple Travelling Salesman Problem AAAI23 天津大学和华为诺亚方舟实验室 摘要 MinMax Multiple Travelling Salesman Problem(mTSP)...

Looking Ahead to Avoid Being Late Solving Hard-Constrained Traveling Salesman Problem

Arxiv

Looking Ahead to Avoid Being Late: Solving Hard-Constrained Traveling Salesman Problem https://arxiv.org/abs/2403.05318v1 (我感觉整体一般) 摘要 许多现实问题都可以被表述为受限旅行商问题(TSP)。然而,约束条件总是复杂和众多的,使得求解tsp具有挑战性。当复杂...