{"id":19484946,"url":"https://github.com/opencog/agents","last_synced_at":"2026-03-01T02:33:55.717Z","repository":{"id":234076067,"uuid":"788251193","full_name":"opencog/agents","owner":"opencog","description":"Experiments with Agents","archived":false,"fork":false,"pushed_at":"2026-02-25T07:10:21.000Z","size":88,"stargazers_count":3,"open_issues_count":0,"forks_count":1,"subscribers_count":5,"default_branch":"master","last_synced_at":"2026-02-25T07:54:28.900Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Scheme","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/opencog.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-04-18T03:59:16.000Z","updated_at":"2026-02-25T07:10:27.000Z","dependencies_parsed_at":"2024-04-18T04:39:41.997Z","dependency_job_id":"d3394db4-d85c-4930-b5e8-755fabadfe3e","html_url":"https://github.com/opencog/agents","commit_stats":null,"previous_names":["opencog/agents"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/opencog/agents","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opencog%2Fagents","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opencog%2Fagents/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opencog%2Fagents/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opencog%2Fagents/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/opencog","download_url":"https://codeload.github.com/opencog/agents/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/opencog%2Fagents/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29959153,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-03-01T01:47:18.291Z","status":"online","status_checked_at":"2026-03-01T02:00:07.437Z","response_time":124,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-10T20:24:50.705Z","updated_at":"2026-03-01T02:33:55.675Z","avatar_url":"https://github.com/opencog.png","language":"Scheme","funding_links":[],"categories":[],"sub_categories":[],"readme":"Atomese Agents\n==============\nExperiments and prototyping for interactive learning agents.\n\n### Status\nVersion 0.0.2 -- Extremely early prototyping stage.\n\n### Overview\nThe [Opencog Learn](https://github.com/opencog/learn) project showed\nthat the basic conscept of structure learning works and is scalable\nand can deal with observing and learning real-world structures. Great!\n\nBut it was built as a batch system, training on large blobs of data,\nmoving from one training stage to the next. This turns out to be\nunmanagable: its complicated, its fragile. It's hard to track status\nand to run experiments. Hard to redo batch steps when bugs get fixed\nand system architecture changes.\n\nMost importantly, the above is a roadblock to scaling: the core idea is\nthat the learning can be done recursively. But if the recursive steps\nteeter precariously on earlier batch runs ... ugh. It became too hard\nto develop the recursive step.\n\nRather than refactoring the code in that git repo, it seems better to\nstart with a clean slate, here. The general ideas get recycled. The\nframework changes dramatically.\n\nOne current goal is to rewrite important parts of the pipeline in pure\nAtomese. This seems like the \"right thing to do\", so that, for example,\nAS-MOSES can be used to mutate and optimize important subsystems: this\nwould be impossible if the processing was done in a conventional\nprogramming language. But these details are unclear, and so all of\nthis is a big experiment. Don't know how it will turn out.\n\nThe hope is that a pure Atomese structure will make the recursive jump\neasy. Without the recursive jump, well, its just Symbolic AI from\nearlier decades.\n\nA general overview can be found in the AGI 2022 paper:\n[Purely Symbolic Induction of Structure](https://github.com/opencog/learn/tree/master/learn-lang-diary/agi-2022/grammar-induction.pdf).\n\nBackground reading: the various PDF and LyX files located at:\n* [AtomSpace Sheaves \u0026 Graphs](https://github.com/opencog/atomspace/tree/master/opencog/sheaf)\n* [OpenCog Learn Project](https://github.com/opencog/learn) and\n  especially the \"diary\" subdirectory there.\n\nAdditional general details and concepts are described in scattershot\nlocations, in diary entries, or nowhere at all.\n\n### Action-Perception\nThe demos here are built on top of the perception-action framework\ndeveloped in the [Sensory project](https://github.com/opencog/sensory).\nThe code here adds some non-trivial processing steps on top of the\nbasic sensori-motor API provided there. Be sure to go through the\nexamples there.\n\nTurns out that action-perception is far more complicated than one might\nthink. Thus, to address the issues discovered there, the\n[OpenCog (Sensori-)Motor project](https://github.com/opencog/motor)\nhas been started. Progress on this project, Agents, is stalled, until\nthe deeper issues explored there are resolved.\n\nFor examples in this project, see the [examples](./examples) directory\nhere.\n\n--------------\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopencog%2Fagents","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fopencog%2Fagents","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fopencog%2Fagents/lists"}