Solving a Tool-Based Interaction Task Using Deep Reinforcement Learning with Visual Attention
Fleer S, Ritter H (2019)
In: Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019. Advances in Intelligent Systems and Computing , 976. Cham: Springer: 231-240.
Sammelwerksbeitrag
| Veröffentlicht | Englisch
Download
Es wurden keine Dateien hochgeladen. Nur Publikationsnachweis!
Autor*in
Einrichtung
Abstract / Bemerkung
We propose a reinforcement learning approach that combines an asynchronous actor-critic model with a recurrent model of visual attention. Instead of using the full visual information of the scene, the resulting model accumulates the foveal information of controlled glimpses and is thus able to reduce the complexity of the network. Using the designed model, an artificial agent is able to solve a challenging “mediated interaction” task. In these tasks, the desired effects cannot be created through direct interaction, but instead require the learner to discover how to exert suitable effects on the target object through involving a “tool”. To learn the given mediated interaction task, the agent is “actively” searching for salient points within the environment by taking a limited number of fovea-like glimpses. It then uses the accumulated information to decide which action to take next.
Stichworte
Reinforcement learning;
Deep learning;
Recurrent neural networks;
Mediated interaction;
Visual attention;
REINFORCE algorithm;
Tool-based interaction
Erscheinungsjahr
2019
Buchtitel
Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019
Serientitel
Advances in Intelligent Systems and Computing
Band
976
Seite(n)
231-240
Konferenz
13th International Workshop on Self-Organizing Maps and Learning Vector Quantization, Clustering and Data Visualization (WSOM)
Konferenzort
Barcelona, Spain
Konferenzdatum
2019-06-26 – 2019-06-28
ISBN
978-3-030-19641-7
eISBN
978-3-030-19642-4
ISSN
2194-5357
Page URI
https://pub.uni-bielefeld.de/record/2935994
Zitieren
Fleer S, Ritter H. Solving a Tool-Based Interaction Task Using Deep Reinforcement Learning with Visual Attention. In: Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019. Advances in Intelligent Systems and Computing . Vol 976. Cham: Springer; 2019: 231-240.
Fleer, S., & Ritter, H. (2019). Solving a Tool-Based Interaction Task Using Deep Reinforcement Learning with Visual Attention. Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019, Advances in Intelligent Systems and Computing , 976, 231-240. Cham: Springer. doi:10.1007/978-3-030-19642-4_23
Fleer, Sascha, and Ritter, Helge. 2019. “Solving a Tool-Based Interaction Task Using Deep Reinforcement Learning with Visual Attention”. In Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019, 976:231-240. Advances in Intelligent Systems and Computing . Cham: Springer.
Fleer, S., and Ritter, H. (2019). “Solving a Tool-Based Interaction Task Using Deep Reinforcement Learning with Visual Attention” in Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019 Advances in Intelligent Systems and Computing , vol. 976, (Cham: Springer), 231-240.
Fleer, S., & Ritter, H., 2019. Solving a Tool-Based Interaction Task Using Deep Reinforcement Learning with Visual Attention. In Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019. Advances in Intelligent Systems and Computing . no.976 Cham: Springer, pp. 231-240.
S. Fleer and H. Ritter, “Solving a Tool-Based Interaction Task Using Deep Reinforcement Learning with Visual Attention”, Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019, Advances in Intelligent Systems and Computing , vol. 976, Cham: Springer, 2019, pp.231-240.
Fleer, S., Ritter, H.: Solving a Tool-Based Interaction Task Using Deep Reinforcement Learning with Visual Attention. Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019. Advances in Intelligent Systems and Computing . 976, p. 231-240. Springer, Cham (2019).
Fleer, Sascha, and Ritter, Helge. “Solving a Tool-Based Interaction Task Using Deep Reinforcement Learning with Visual Attention”. Advances in Self-Organizing Maps, Learning Vector Quantization, Clustering and Data Visualization. WSOM 2019. Cham: Springer, 2019.Vol. 976. Advances in Intelligent Systems and Computing . 231-240.
Link(s) zu Volltext(en)
Access Level
Closed Access
Material in PUB:
Teil von PUB Eintrag
Supplementary Material - Solving a tool-based interaction task using deep reinforcement learning with visual attention
Fleer S (2019)
Bielefeld University.
Fleer S (2019)
Bielefeld University.
Google Scholar
ISBN Suche