WEKO3
アイテム
Emergence of Content-Agnostic Information Processing by a Robot Using Active Inference, Visual Attention, Working Memory, and Planning
https://oist.repo.nii.ac.jp/records/2333
https://oist.repo.nii.ac.jp/records/233383a582c3-9ec5-43ad-aea5-a23b98fd2610
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
Queisser-2021-Emergence of Content-Agnostic In_VoR (2.3 MB)
|
|
| Item type | 学術雑誌論文 / Journal Article(1) | |||||
|---|---|---|---|---|---|---|
| 公開日 | 2021-10-18 | |||||
| タイトル | ||||||
| タイトル | Emergence of Content-Agnostic Information Processing by a Robot Using Active Inference, Visual Attention, Working Memory, and Planning | |||||
| 言語 | en | |||||
| 言語 | ||||||
| 言語 | eng | |||||
| 資源タイプ | ||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_6501 | |||||
| 資源タイプ | journal article | |||||
| 著者（英） |
Queiẞer, Jeffrey Frederic
× Queiẞer, Jeffrey Frederic× Jung, Minju× Matsumoto, Takazumi× Tani, Jun |
|||||
| 書誌情報 |
en : Neural Computation 巻 33, 号 9, p. 2353-2407, 発行日 2021-08-19 |
|||||
| 抄録 | ||||||
| 内容記述タイプ | Other | |||||
| 内容記述 | Generalization by learning is an essential cognitive competency for humans. For example, we can manipulate even unfamiliar objects and can generate mental images before enacting a preplan. How is this possible? Our study investigated this problem by revisiting our previous study (Jung, Matsumoto, & Tani, 2019), which examined the problem of vision-based, goal-directed planning by robots performing a task of block stacking. By extending the previous study, our work introduces a large network comprising dynamically interacting submodules, including visual working memory (VWMs), a visual attention module, and an executive network. The executive network predicts motor signals, visual images, and various controls for attention, as well as masking of visual information. The most significant difference from the previous study is that our current model contains an additional VWM. The entire network is trained by using predictive coding and an optimal visuomotor plan to achieve a given goal state is inferred using active inference. Results indicate that our current model performs significantly better than that used in Jung et al. (2019), especially when manipulating blocks with unlearned colors and textures. Simulation results revealed that the observed generalization was achieved because content-agnostic information processing developed through synergistic interaction between the second VWM and other modules during the course of learning, in which memorizing image contents and transforming them are dissociated. This letter verifies this claim by conducting both qualitative and quantitative analysis of simulation results. | |||||
| 出版者 | ||||||
| 出版者 | MIT Press | |||||
| ISSN | ||||||
| 収録物識別子タイプ | ISSN | |||||
| 収録物識別子 | 0899-7667 | |||||
| ISSN | ||||||
| 収録物識別子タイプ | ISSN | |||||
| 収録物識別子 | 1530-888X | |||||
| PubMed番号 | ||||||
| 関連タイプ | isIdenticalTo | |||||
| 識別子タイプ | PMID | |||||
| 関連識別子 | info:pmid/34412116 | |||||
| DOI | ||||||
| 関連タイプ | isIdenticalTo | |||||
| 識別子タイプ | DOI | |||||
| 関連識別子 | info:doi/10.1162/neco_a_01412 | |||||
| 権利 | ||||||
| 権利情報 | © 2021 Massachusetts Institute of Technology | |||||
| 関連サイト | ||||||
| 識別子タイプ | URI | |||||
| 関連識別子 | https://direct.mit.edu/neco/article/33/9/2353/102624/Emergence-of-Content-Agnostic-Information | |||||
| 著者版フラグ | ||||||
| 出版タイプ | VoR | |||||
| 出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||
