[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-11430":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":15,"stars7d":17,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":10,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":23,"hasPages":23,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":16,"starSnapshotCount":16,"syncStatus":14,"lastSyncTime":28,"discoverSource":29},11430,"learning-beyond-gradients","Trinkle23897\u002Flearning-beyond-gradients","Trinkle23897","Heuristic Learning Blog Post","https:\u002F\u002Ftrinkle23897.github.io\u002Flearning-beyond-gradients\u002F",null,"Python",554,57,2,15,0,25,169,45,9.29,false,"main",true,[],"2026-06-12 02:02:31","# Learning Beyond Gradients Blog Artifacts\n\nThis repository contains the public artifacts for:\n\n**Learning Beyond Gradients**\n\nPublished article:\n\n- https:\u002F\u002Ftrinkle23897.github.io\u002Flearning-beyond-gradients\u002F\n\nArtifact repository:\n\n- https:\u002F\u002Fgithub.com\u002FTrinkle23897\u002Flearning-beyond-gradients\n\nThe article is bilingual. The rendered HTML defaults to English and includes a Chinese switcher.\n\n## Source Files\n\n- `learning-beyond-gradient.en.md`: English article source.\n- `learning-beyond-gradient.md`: Chinese article source.\n- `learning-beyond-gradient.html`: rendered bilingual HTML.\n- `render_learning_beyond_gradient.py`: local renderer.\n\nThe deployed article is `learning-beyond-gradient.html`.\n\n## Local Preview\n\nFrom the repository root:\n\n```bash\npython3 -m http.server 8000\n```\n\nThen open:\n\n```text\nhttp:\u002F\u002F127.0.0.1:8000\u002Flearning-beyond-gradient.html\n```\n\nOpening the HTML file directly also works in most browsers, but using `http.server` is closer to how the page is served.\n\n## Re-render The HTML\n\nInstall the rendering dependency:\n\n```bash\npython3 -m pip install -r requirements.txt\n```\n\nThen run:\n\n```bash\npython3 render_learning_beyond_gradient.py\n```\n\nThe renderer reads the English and Chinese Markdown files and rewrites `learning-beyond-gradient.html` in place.\n\n## GitHub Pages\n\nThe site is deployed by `.github\u002Fworkflows\u002Fdeploy-pages.yml` on every push to `main`.\n\nThe workflow does not publish the whole repository as the website root. It builds a small `_site` directory containing:\n\n- `index.html`, copied from `learning-beyond-gradient.html`.\n- `.nojekyll`.\n- Local files referenced by the article through `src` or `href`, such as figures, videos, scripts, CSVs, and prompt files.\n\n## Included Artifacts\n\nThe repository includes the files needed to inspect and reproduce the article's representative results:\n\n- `atari\u002Fpong\u002F`: Pong policy script.\n- `atari\u002Fbreakout\u002F`: Breakout policy, trial summaries, sample-efficiency figure, and checkpoint videos.\n- `atari\u002Fmontezuma\u002F`: Montezuma exploratory policies, state\u002Farchive search scripts, summaries, probe images, and replay artifacts.\n- `atari\u002Fatari57\u002F`: Atari57 aggregate\u002Fper-game figures, CSV summaries, and the batch prompt template used for unattended Codex CLI runs.\n- `mujoco\u002Fant\u002F`: Ant policy, minimal extracted Ant policy, trial summaries, MuJoCo XML, sample-efficiency figure, and final-policy video.\n- `mujoco\u002Fhalfcheetah\u002F`: HalfCheetah policy script, iteration log, and sample-efficiency figure.\n- `vizdoom\u002F`: D1\u002FD3 VizDoom heuristic scripts plus 35fps 10-seed render videos.\n\nThe article appendix contains reproduction commands for several representative results. Those commands assume they are run from the repository root after cloning this repo.\n\n## Runtime Notes\n\nThe experiments were written against EnvPool `1.1.1`. The article commands assume the relevant Python environment already has EnvPool and the Atari\u002FMuJoCo runtime dependencies installed.\n\nFor Ant, `ant_envpool.xml` stays next to `heuristic_ant.py` under `mujoco\u002Fant\u002F`. The reproduction command references it as:\n\n```bash\n--mujoco-xml-path mujoco\u002Fant\u002Fant_envpool.xml\n```\n\n## Citation\n\n```bibtex\n@misc{weng2026learning_beyond_gradients,\n  title = {Learning Beyond Gradients},\n  author = {Weng, Jiayi},\n  year = {2026},\n  month = may,\n  howpublished = {\\url{https:\u002F\u002Ftrinkle23897.github.io\u002Flearning-beyond-gradients\u002F}},\n  note = {Blog post}\n}\n```\n","该项目旨在提供关于超越梯度学习的研究文章及其相关资源。它包含了一篇双语（中英文）的文章，以及用于重现文章中提到实验结果的数据集和脚本。技术上，项目使用Python编写，并通过一个简单的HTTP服务器进行本地预览。此外，还提供了渲染工具来更新HTML版本的文章内容。适合对深度强化学习、特别是对探索除传统梯度方法外其他学习策略感兴趣的科研人员或开发者参考使用。","2026-06-11 03:31:51","CREATED_QUERY"]