December 4, 2018 Software Open Access

Ivan Sosin; Oleg Svidchenko; Aleksandra Malysheva; Daniel Kudenko; Aleksei Shpilman

JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/197c74f2-a668-43a7-8f46-2635582f63d1/iasawseen/MultiServerRL-v1.0.zip"
      }, 
      "checksum": "md5:569c93638c7a3485b8c74839b7f16af0", 
      "bucket": "197c74f2-a668-43a7-8f46-2635582f63d1", 
      "key": "iasawseen/MultiServerRL-v1.0.zip", 
      "type": "zip", 
      "size": 20187
    }
  ], 
  "owners": [
    56783
  ], 
  "doi": "10.5281/zenodo.1938263", 
  "stats": {
    "version_unique_downloads": 17.0, 
    "unique_views": 387.0, 
    "views": 414.0, 
    "version_views": 411.0, 
    "unique_downloads": 17.0, 
    "version_unique_views": 385.0, 
    "volume": 343179.0, 
    "version_downloads": 17.0, 
    "downloads": 17.0, 
    "version_volume": 343179.0
  }, 
  "links": {
    "doi": "https://doi.org/10.5281/zenodo.1938263", 
    "conceptdoi": "https://doi.org/10.5281/zenodo.1938262", 
    "bucket": "https://zenodo.org/api/files/197c74f2-a668-43a7-8f46-2635582f63d1", 
    "conceptbadge": "https://zenodo.org/badge/doi/10.5281/zenodo.1938262.svg", 
    "html": "https://zenodo.org/record/1938263", 
    "latest_html": "https://zenodo.org/record/1938263", 
    "badge": "https://zenodo.org/badge/doi/10.5281/zenodo.1938263.svg", 
    "latest": "https://zenodo.org/api/records/1938263"
  }, 
  "conceptdoi": "10.5281/zenodo.1938262", 
  "created": "2018-12-04T19:12:42.840891+00:00", 
  "updated": "2020-01-25T19:22:14.197102+00:00", 
  "conceptrecid": "1938262", 
  "revision": 7, 
  "id": 1938263, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.5281/zenodo.1938263", 
    "description": "<p>One of the main challenges faced in Deep Reinforcement Learning is that running simulations may be CPU-heavy, while the optimal computing device for training neural networks is a GPU. One way to overcome this problem is building a custom machine with GPU to CPU proportions that avoid bottlenecking one or the other. Another is to have the GPU machine work together with the CPU machine and/or launching one or both via cloud computing service. We have designed a framework for such a tandem interaction.</p>\n\n<p>Authors: Ivan Sosin, Oleg Svidchenko, Aleksandra Malysheva, Daniel Kudenko, Aleksei Shpilman.</p>", 
    "license": {
      "id": "other-open"
    }, 
    "title": "Framework for Deep Reinforcement Learning with GPU-CPU Multiprocessing", 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "1938262"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "1938263"
          }
        }
      ]
    }, 
    "version": "v1.0", 
    "publication_date": "2018-12-04", 
    "creators": [
      {
        "affiliation": "JetBrains Research", 
        "name": "Ivan Sosin"
      }, 
      {
        "affiliation": "JetBrains Research", 
        "name": "Oleg Svidchenko"
      }, 
      {
        "affiliation": "JetBrains Research", 
        "name": "Aleksandra Malysheva"
      }, 
      {
        "affiliation": "JetBrains Research", 
        "name": "Daniel Kudenko"
      }, 
      {
        "affiliation": "JetBrains Research", 
        "name": "Aleksei Shpilman"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "type": "software", 
      "title": "Software"
    }, 
    "related_identifiers": [
      {
        "scheme": "url", 
        "identifier": "https://github.com/iasawseen/MultiServerRL/tree/v1.0", 
        "relation": "isSupplementTo"
      }, 
      {
        "scheme": "doi", 
        "identifier": "10.5281/zenodo.1938262", 
        "relation": "isVersionOf"
      }
    ]
  }
}