The field of optimization is witnessing significant developments, with a focus on solving complex problems efficiently. Researchers are exploring new approaches to tackle large-scale optimization challenges, including the use of graph filters, neural networks, and reinforcement learning. These innovative methods are enabling the solution of previously intractable problems, such as partially observable Markov decision processes and capacitated arc routing problems. Noteworthy papers in this area include: Partially Observable Monte-Carlo Graph Search, which proposes a new sampling-based algorithm for solving large POMDPs offline. Efficient Neural Combinatorial Optimization Solver for the Min-max Heterogeneous Capacitated Vehicle Routing Problem, which introduces an efficient NCO solver that outperforms state-of-the-art solvers. Unrolling Dynamic Programming via Graph Filters, which proposes a new approach that unrolls and truncates policy iterations into a learnable parametric model. Knowledge-Guided Memetic Algorithm for Capacitated Arc Routing Problems with Time-Dependent Service Costs, which proposes a knowledge-guided memetic algorithm that achieves higher search efficiency than state-of-the-art methods. Spatial-Temporal Reinforcement Learning for Network Routing with Non-Markovian Traffic, which proposes a spatial-temporal RL approach that integrates Graph Neural Networks and Recurrent Neural Networks. Parametrized Multi-Agent Routing via Deep Attention Models, which proposes a scalable deep learning framework for parametrized sequential decision-making. Nearest-Better Network for Visualizing and Analyzing Combinatorial Optimization Problems, which provides a straightforward theoretical derivation and an efficient NBN computation method. Inside madupite: Technical Design and Performance, which introduces and benchmarks a newly proposed high-performance solver designed for large-scale discounted infinite-horizon Markov decision processes.