Machine learning has been adapted to help solve NP-hard combinatorial
optimization problems. One prevalent way is learning to construct solutions by
deep neural networks, which has been receiving more and more attention due to
the high efficiency and less requirement for expert knowledge. However, many
neural construction methods for Vehicle Routing Problems (VRPs) focus on
synthetic problem instances with limited scales and specified node
distributions, leading to poor performance on real-world problems which usually
involve large scales together with complex and unknown node distributions. To
make neural VRP solvers more practical in real-world scenarios, we design an
auxiliary policy that learns from the local transferable topological features,
named local policy, and integrate it with a typical constructive policy (which
learns from the global information of VRP instances) to form an ensemble
policy. With joint training, the aggregated policies perform cooperatively and
complementarily to boost generalization. The experimental results on two
well-known benchmarks, TSPLIB and CVRPLIB, of travelling salesman problem and
capacitated VRP show that the ensemble policy consistently achieves better
generalization than state-of-the-art construction methods and even works well
on real-world problems with several thousand nodes.

为了提高神经车辆路径问题（VRP）求解器在实际环境中的可行性，我们设计了一个辅助策略，通过学习本地可转移的拓扑特征来改善神经 VRP 求解器的表现，将其与典型的建构策略相结合，形成一个集成策略。经过联合训练，这个集成策略使得各个策略相互协作和互补，从而提高了泛化能力。实验证明，与最先进的构造方法相比，这个集成策略在两个著名的基准测试中（TSPLIB 和 CVRPLIB）表现出更好的泛化性能，并且在数千个节点的真实世界问题上也能工作得很好。

通过可传递的本地策略集成实现面向可广泛应用的神经求解器车辆路径问题

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

Serverless cloud is an innovative cloud service model that frees customers
from most cloud management duties. It also offers the same advantages as other
cloud models but at much lower costs. As a result, the serverless cloud has
been increasingly employed in high-impact areas such as system security,
banking, and health care. A big threat to the serverless cloud's performance is
cold-start, which is when the time of provisioning the needed cloud resource to
serve customers' requests incurs unacceptable costs to the service providers
and/or the customers. This paper proposes a novel low-coupling, high-cohesion
ensemble policy that addresses the cold-start problem at infrastructure- and
function-levels of the serverless cloud stack, while the state of the art
policies have a more narrowed focus. This ensemble policy anchors on the
prediction of function instance arrivals, 10 to 15 minutes into the future. It
is achievable by using the temporal convolutional network (TCN) deep-learning
method. Bench-marking results on a real-world dataset from a large-scale
serverless cloud provider show that TCN out-performs other popular machine
learning algorithms for time series. Going beyond cold-start management, the
proposed policy and publicly available codes can be adopted in solving other
cloud problems such as optimizing the provisioning of virtual software-defined
network assets.

该论文提出了一种新的低耦合、高内聚的集成策略以解决无服务器云堆栈基础结构和功能级别的冷启动问题，并提出了一种基于时间卷积网络（TCN）的深度学习方法，用于预测未来 10 至 15 分钟内功能实例的到来，并在实际数据集上进行了基准测试，证明 TCN 在时序机器学习算法中表现出更好的性能。