[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78VQD8SPBNP1043NJEP2WM6":3,"skill-same:01M78VQD8SPBNP1043NJEP2WM6":62},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":32,"install":33,"resources":45},"01M78VQD8SPBNP1043NJEP2WM6","value-based-discrete-control",null,"Use the repository's Char01 DQN family for discrete Gym control: CartPole, MountainCar, replay buffers, reward shaping, target-network updates, and TensorBoard logging.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"MIT","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-reinforcement-learning-with-pytorch\u002Fsub-skills\u002Fvalue-based-discrete-control\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: value-based-discrete-control\ndescription: \"Use the repository's Char01 DQN family for discrete Gym control:\n  CartPole, MountainCar, replay buffers, reward shaping, target-network updates,\n  and TensorBoard logging.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: MIT\n---\n\n# Value-Based Discrete Control\n\nUse this sub-skill for requests about the repository's Deep Q-Network examples in\nclassic discrete-action Gym environments. It is a routing and operating guide for\nthe Char01 DQN family, not a packaged training API.\n\n## Use when\n\n- The user asks to run, adapt, or explain the CartPole DQN workflow.\n- The user asks why MountainCar DQN is hard, how to shape its reward, or how to\n  tune its replay\u002Fupdate settings.\n- The user asks about the DQN replay buffer layout, sampling behavior,\n  epsilon-greedy action selection, target network copy cadence, or TensorBoard\n  loss\u002Fepisode logs.\n- The user needs a bounded compatibility probe for Gym classic-control + torch +\n  tensorboardX before adapting a DQN script.\n\n## Route elsewhere\n\n- Tabular Q-learning\u002FSarsa\u002FGridWorld requests -> sibling `tabular-control`.\n- REINFORCE, policy-gradient, actor-critic, A2C, or PPO requests -> sibling\n  `on-policy-actor-critic`.\n- DDPG, SAC, TD3, Pendulum, BipedalWalker, or continuous-action off-policy\n  requests -> sibling `off-policy-continuous-control`.\n- Plot-only curve aggregation requests -> root\u002Fshared plotting support.\n\n## Runtime entry points\n\n1. For a DQN-family decision or code adaptation, read\n   `references\u002Fdqn-family-map.md` first.\n2. For failures, surprising learning curves, headless runs, TensorBoard output,\n   or Gym-version issues, read `references\u002Ftroubleshooting.md`.\n3. For a bounded environment\u002FAPI check, run the bundled inspection helper rather\n   than launching a long training loop:\n\n   ```bash\n   python sub-skills\u002Fvalue-based-discrete-control\u002Fscripts\u002Fdqn_discrete_probe.py --env CartPole-v0 --steps 3\n   python sub-skills\u002Fvalue-based-discrete-control\u002Fscripts\u002Fdqn_discrete_probe.py --env MountainCar-v0 --steps 3 --reward-shaping mountaincar-position-bonus\n   ```\n\n   The helper performs no learning and never calls `render()`.\n\n## Source-script policy\n\nThe repository's DQN scripts are training-scale references, not safe default\nhelpers. Treat them as distilled patterns:\n\n- `DQN_CartPole-v0.py` and `DQN_MountainCar-v0.py`: list replay buffer,\n  tensorboardX logging, target\u002Fact networks, long episode counts.\n- `DQN.py`, `naiveDQN.py`, and `DQN_mountain_car_v1.py`: older numpy-ring-buffer\n  variants with unconditional render\u002Fplot behavior; use for algorithm notes and\n  reward-shaping clues only.\n\nDo not run the original long loops by default. If a user explicitly asks to run\ntraining, first make it bounded: disable rendering, choose a writable logdir,\nreduce episodes\u002Fcapacity for a smoke, and preserve old-Gym reset\u002Fstep semantics\nor add a compatibility wrapper.\n\n## Fast operating checklist\n\n- Identify env family: CartPole has dense-ish balance reward; MountainCar needs\n  explicit discussion of sparse\u002Fstep-penalty learning and shaping.\n- Identify replay style: list replay (`capacity=8000`, `batch_size=256`) versus\n  numpy ring replay (`MEMORY_CAPACITY=2000` or `20000`).\n- Check learning gate: most variants do not update until replay memory is full.\n- Check target sync cadence: target networks are hard-copied every 100 update or\n  learn steps, depending on the variant.\n- Keep runtime edits self-contained; do not depend on the original checkout.\n","# Value-Based Discrete Control\n\nUse this sub-skill for requests about the repository's Deep Q-Network examples in\nclassic discrete-action Gym environments. It is a routing and operating guide for\nthe Char01 DQN family, not a packaged training API.\n\n## Use when\n\n- The user asks to run, adapt, or explain the CartPole DQN workflow.\n- The user asks why MountainCar DQN is hard, how to shape its reward, or how to\n  tune its replay\u002Fupdate settings.\n- The user asks about the DQN replay buffer layout, sampling behavior,\n  epsilon-greedy action selection, target network copy cadence, or TensorBoard\n  loss\u002Fepisode logs.\n- The user needs a bounded compatibility probe for Gym classic-control + torch +\n  tensorboardX before adapting a DQN script.\n\n## Route elsewhere\n\n- Tabular Q-learning\u002FSarsa\u002FGridWorld requests -> sibling `tabular-control`.\n- REINFORCE, policy-gradient, actor-critic, A2C, or PPO requests -> sibling\n  `on-policy-actor-critic`.\n- DDPG, SAC, TD3, Pendulum, BipedalWalker, or continuous-action off-policy\n  requests -> sibling `off-policy-continuous-control`.\n- Plot-only curve aggregation requests -> root\u002Fshared plotting support.\n\n## Runtime entry points\n\n1. For a DQN-family decision or code adaptation, read\n   `references\u002Fdqn-family-map.md` first.\n2. For failures, surprising learning curves, headless runs, TensorBoard output,\n   or Gym-version issues, read `references\u002Ftroubleshooting.md`.\n3. For a bounded environment\u002FAPI check, run the bundled inspection helper rather\n   than launching a long training loop:\n\n   ```bash\n   python sub-skills\u002Fvalue-based-discrete-control\u002Fscripts\u002Fdqn_discrete_probe.py --env CartPole-v0 --steps 3\n   python sub-skills\u002Fvalue-based-discrete-control\u002Fscripts\u002Fdqn_discrete_probe.py --env MountainCar-v0 --steps 3 --reward-shaping mountaincar-position-bonus\n   ```\n\n   The helper performs no learning and never calls `render()`.\n\n## Source-script policy\n\nThe repository's DQN scripts are training-scale references, not safe default\nhelpers. Treat them as distilled patterns:\n\n- `DQN_CartPole-v0.py` and `DQN_MountainCar-v0.py`: list replay buffer,\n  tensorboardX logging, target\u002Fact networks, long episode counts.\n- `DQN.py`, `naiveDQN.py`, and `DQN_mountain_car_v1.py`: older numpy-ring-buffer\n  variants with unconditional render\u002Fplot behavior; use for algorithm notes and\n  reward-shaping clues only.\n\nDo not run the original long loops by default. If a user explicitly asks to run\ntraining, first make it bounded: disable rendering, choose a writable logdir,\nreduce episodes\u002Fcapacity for a smoke, and preserve old-Gym reset\u002Fstep semantics\nor add a compatibility wrapper.\n\n## Fast operating checklist\n\n- Identify env family: CartPole has dense-ish balance reward; MountainCar needs\n  explicit discussion of sparse\u002Fstep-penalty learning and shaping.\n- Identify replay style: list replay (`capacity=8000`, `batch_size=256`) versus\n  numpy ring replay (`MEMORY_CAPACITY=2000` or `20000`).\n- Check learning gate: most variants do not update until replay memory is full.\n- Check target sync cadence: target networks are hard-copied every 100 update or\n  learn steps, depending on the variant.\n- Keep runtime edits self-contained; do not depend on the original checkout.\n",{"name":5,"license":14,"metadata":23,"description":7,"disable-model-invocation":25},{"disco-role":24},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fdeep-reinforcement-learning-with-pytorch\u002Fsub-skills\u002Fvalue-based-discrete-control",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":25},[35,38,41],{"tool":36,"command":37},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill value-based-discrete-control",{"tool":39,"command":40},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill value-based-discrete-control",{"tool":42,"command":43},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill value-based-discrete-control","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-reinforcement-learning-with-pytorch\u002Fsub-skills\u002Fvalue-based-discrete-control\u002FSKILL.md",[46,52,56],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fdqn-family-map.md","REFERENCE","参考文档",6150,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-reinforcement-learning-with-pytorch\u002Fsub-skills\u002Fvalue-based-discrete-control\u002Freferences\u002Fdqn-family-map.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Ftroubleshooting.md",3444,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-reinforcement-learning-with-pytorch\u002Fsub-skills\u002Fvalue-based-discrete-control\u002Freferences\u002Ftroubleshooting.md",{"path":57,"resourceType":58,"resourceTypeDesc":59,"sizeBytes":60,"sourceHtmlUrl":61},"scripts\u002Fdqn_discrete_probe.py","SCRIPT","脚本",6190,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fdeep-reinforcement-learning-with-pytorch\u002Fsub-skills\u002Fvalue-based-discrete-control\u002Fscripts\u002Fdqn_discrete_probe.py",[63,71,78,85,92,99],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":18},"01M799NYSH1R3RVAHSM2D88Y84","passnet-triton-opt","Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block\u002Fgrid\u002Fwarp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":18},"01M799NS5APEC0N1FZ7XZVW6BN","passnet-solve","END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-solve\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":18},"01M799NNY9BKSM21V5R7QVTBJX","passnet-skill","PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under .\u002Fpass_dir\u002F, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-skill\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":18},"01M799NHH6NP7Z1RK77KGXAN6S","passnet-pattern-fusion","Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()\u002Freplacement_args()\u002Freplacement_func() files, when a pass \"failed to match\", or when deciding how to fuse multiple kernels\u002Fops.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-pattern-fusion\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":18},"01M799ND2J4WNC2PGAZ0A06NDF","passnet-orchestrate","Round\u002Fbatch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep\u002Frevert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-orchestrate\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":67,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":18},"01M799N8SBTMY02ZDG2AZF5ZWE","passnet-feedback","Fast, deterministic feedback for PassNet work: pre-flight pattern\u002Fmatch verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-feedback\u002FSKILL.md",[],[]]