<![CDATA[panbaoran913的博客]]>

<![CDATA[panbaoran913的博客]]>https://blog.youkuaiyun.com/panbaoran913zh-cnhttps://blog.youkuaiyun.com/<![CDATA[[论] Optimal Design of Observable Multi-Agent Networks: A Structural System Approach]]>https://blog.youkuaiyun.com/panbaoran913/article/details/158890662https://blog.youkuaiyun.com/panbaoran913/article/details/158890662panbaoran913Tue, 10 Mar 2026 22:02:07 +0800<![CDATA[【专题之MARL】 Prescriptive Process Monitoring Under Resource Constraints: A Reinforcement Learning Appro]]>https://blog.youkuaiyun.com/panbaoran913/article/details/158419181https://blog.youkuaiyun.com/panbaoran913/article/details/158419181panbaoran913Fri, 27 Feb 2026 13:58:52 +0800<![CDATA[【专题之MARL】 A Multi-Agent Reinforcement Learning Approach to Prescriptive Process Monitoring]]>https://blog.youkuaiyun.com/panbaoran913/article/details/158347772https://blog.youkuaiyun.com/panbaoran913/article/details/158347772panbaoran913Wed, 25 Feb 2026 11:42:00 +0800<![CDATA[【专题之MARL】A Survey on UAV Control with Multi-Agent Reinforcement Learning]]>https://blog.youkuaiyun.com/panbaoran913/article/details/158316045https://blog.youkuaiyun.com/panbaoran913/article/details/158316045panbaoran913Mon, 23 Feb 2026 21:20:19 +0800<![CDATA[【报错】import导入警告但实际已经安装]]>https://blog.youkuaiyun.com/panbaoran913/article/details/156569541https://blog.youkuaiyun.com/panbaoran913/article/details/156569541panbaoran913Sun, 04 Jan 2026 15:44:18 +0800"Vritualenv环境“–> “现有”。因为这个包是下载的且有设定配置的，并非是pycharm生成全新环境。在解释器一栏：选择D:\env……这种配置喜好文件删除。]]><![CDATA[【上海(2)】自行车数据协同路网、天气、区级数据解决问题--新手小白保姆级]]>https://blog.youkuaiyun.com/panbaoran913/article/details/156084482https://blog.youkuaiyun.com/panbaoran913/article/details/156084482panbaoran913Thu, 25 Dec 2025 19:36:29 +0800<![CDATA[【上海(1)】自行车数据协同路网、天气、区级数据解决问题--新手小白保姆级]]>https://blog.youkuaiyun.com/panbaoran913/article/details/156076295https://blog.youkuaiyun.com/panbaoran913/article/details/156076295panbaoran913Fri, 19 Dec 2025 12:09:18 +0800<![CDATA[【综精】Dynamic rebalancing optimization for bike-sharing systems: A modeling framework and empirical c]]>https://blog.youkuaiyun.com/panbaoran913/article/details/152086432https://blog.youkuaiyun.com/panbaoran913/article/details/152086432panbaoran913Wed, 12 Nov 2025 11:14:05 +0800<![CDATA[【精析】A Reinforcement Learning Approach for Dynamic Rebalancing in Bike-Sharing Systems]]>https://blog.youkuaiyun.com/panbaoran913/article/details/151405053https://blog.youkuaiyun.com/panbaoran913/article/details/151405053panbaoran913Thu, 25 Sep 2025 16:22:08 +0800<![CDATA[【简】深度强化学习理论及其应用综述]]>https://blog.youkuaiyun.com/panbaoran913/article/details/151834717https://blog.youkuaiyun.com/panbaoran913/article/details/151834717panbaoran913Thu, 18 Sep 2025 17:28:42 +0800<![CDATA[【综析】强化学习稀疏奖励算法研究——理论与实验]]>https://blog.youkuaiyun.com/panbaoran913/article/details/151766723https://blog.youkuaiyun.com/panbaoran913/article/details/151766723panbaoran913Thu, 18 Sep 2025 00:43:55 +0800<![CDATA[DQN在稀疏奖励中的局限性]]>https://blog.youkuaiyun.com/panbaoran913/article/details/151766377https://blog.youkuaiyun.com/panbaoran913/article/details/151766377panbaoran913Tue, 16 Sep 2025 23:07:10 +0800<![CDATA[DQN内存泄露]]>https://blog.youkuaiyun.com/panbaoran913/article/details/151759302https://blog.youkuaiyun.com/panbaoran913/article/details/151759302panbaoran913Tue, 16 Sep 2025 18:51:21 +0800<![CDATA[[旧]Markov Decision Processes: A Tool for Sequential Decision Making under Uncertainty]]>https://blog.youkuaiyun.com/panbaoran913/article/details/151328592https://blog.youkuaiyun.com/panbaoran913/article/details/151328592panbaoran913Mon, 08 Sep 2025 19:13:54 +0800<![CDATA[[精析]Dual Policy Reinforcement Learning for Real-time Rebalancing in Bike-sharing Systems]]>https://blog.youkuaiyun.com/panbaoran913/article/details/150929972https://blog.youkuaiyun.com/panbaoran913/article/details/150929972panbaoran913Thu, 28 Aug 2025 01:51:49 +0800<![CDATA[【论】Trajectory Forecasts in Unknown Environments Conditioned on Grid-Based Plans]]>https://blog.youkuaiyun.com/panbaoran913/article/details/150881293https://blog.youkuaiyun.com/panbaoran913/article/details/150881293panbaoran913Wed, 27 Aug 2025 00:49:22 +0800<![CDATA[【普】人机协同决策的异质多智能体路径规划]]>https://blog.youkuaiyun.com/panbaoran913/article/details/149472641https://blog.youkuaiyun.com/panbaoran913/article/details/149472641panbaoran913Sun, 20 Jul 2025 01:17:23 +0800<![CDATA[【精】Generating Adjacency-Constrained Subgoals in Hierarchical Reinforcement Learning]]>https://blog.youkuaiyun.com/panbaoran913/article/details/149442796https://blog.youkuaiyun.com/panbaoran913/article/details/149442796panbaoran913Fri, 18 Jul 2025 14:14:59 +0800<![CDATA[【析精】Landmark-Guided Subgoal Generation in Hierarchical Reinforcement Learning]]>https://blog.youkuaiyun.com/panbaoran913/article/details/149409038https://blog.youkuaiyun.com/panbaoran913/article/details/149409038panbaoran913Thu, 17 Jul 2025 18:49:27 +0800<![CDATA[[析]Deep reinforcement learning for drone navigation using sensor data]]>https://blog.youkuaiyun.com/panbaoran913/article/details/149370281https://blog.youkuaiyun.com/panbaoran913/article/details/149370281panbaoran913Wed, 16 Jul 2025 22:35:17 +080032.使用了PPO8, PPO16, PPO，和启发式算法(基准)作对比实验。PPO8和PPO16在训练初期学习速度较慢，但后期表现更稳定。随着训练进行，PPO8和PPO16的性能逐渐超越PPO，表明记忆机制的有效性。 LSTM的时间步数为4or8。输入一个长度为n的状态序列。每步的状态为[时间，障碍物信息，距离dx, dy]。LSTM输出的是策略(即【动作，概率】)，LSTM的优化不]]>