https://junfei-z.github.io/tags/networked-systems/Networked SystemsHugo Blox Builder (https://hugoblox.com)en-usMon, 17 Mar 2025 00:00:00 +0000https://junfei-z.github.io/media/icon_hu70bcee51a3cd7a7338014254a2e0c844_1401285_512x512_fill_lanczos_center_3.pnghttps://junfei-z.github.io/tags/networked-systems/https://junfei-z.github.io/project/2_stock/Mon, 17 Mar 2025 00:00:00 +0000https://junfei-z.github.io/project/2_stock/<p>Bridged deterministic optimal control and reinforcement learning to develop a stochastic vaccine allocation strategy on individual-level contact networks, enabling robust pandemic response modeling.</p> <h2 id="highlights">Highlights</h2> <ul> <li>Modeled epidemic spread using a high-dimensional continuous-time Markov process (CTMP) on a contact graph.</li> <li>Designed a vaccination policy using policy gradient-based RL, warm-started from a mean-field ODE solution.</li> <li>Evaluated policies on metrics like mortality and hospitalizations across synthetic and real-world network topologies.</li> </ul> <h2 id="tools">Tools</h2> <p>Python, PyTorch, NetworkX, OpenAI Gym</p>https://junfei-z.github.io/zh/project/2_stock/Mon, 17 Mar 2025 00:00:00 +0000https://junfei-z.github.io/zh/project/2_stock/<p>将确定性最优控制与 Reinforcement Learning 相结合，开发了个体级接触网络上的随机疫苗分配策略，实现了鲁棒的疫情响应建模。</p> <h2 id="项目亮点">项目亮点</h2> <ul> <li>在接触图上使用高维连续时间马尔可夫过程 (CTMP) 对疫情传播进行建模。</li> <li>设计了基于 Policy Gradient 的 RL 疫苗接种策略，并以 Mean-Field ODE 解作为热启动。</li> <li>在合成和真实世界网络拓扑上评估了策略在死亡率和住院率等指标上的表现。</li> </ul> <h2 id="工具">工具</h2> <p>Python, PyTorch, NetworkX, OpenAI Gym</p>