Figure 7 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Download
View publication
Possible actions of an agent in a PCN environment. At each location, the agent can take 12 possible actions. Depending on the action, the agent can remain on the same photonic chip or move to an adjacent one.

Possible actions of an agent in a PCN environment. At each location, the agent can take 12 possible actions. Depending on the action, the agent can remain on the same photonic chip or move to an adjacent one.

Source publication
Figure 2: The architecture of our proposed four-degree 3×3×3×3 silicon...
Figure 5: A typical routing problem in a Photonic Chip Network (PCN),...
Figure 6: Markov Decision Process in PCN environment. S T and S T +1...
Figure 7: Possible actions of an agent in a PCN environment. At each...
+4 Figure 8: Learning the episode that corresponds to a failed routing....
Self-Controlling Photonic-on-Chip Networks With Deep Reinforcement Learning
Preprint
Full-text available
  • Jul 2021
We present a novel photonic chip design for high bandwidth four-degree optical switches that support high-dimensional switching mechanisms with low insertion loss and low crosstalk in a low power consumption level and a short switching time. Such four-degree photonic chips can be used to build an integrated full-grid Photonic-on-Chip Network (PCN)....
Cite
Download full-text

Context in source publication

Context 1
... on the action, the agent can remain on the same chip or move to an adjacent chip. For example, consider a specific agent at port O2 (Figure 7). This agent will remain in the same chip if it takes any Up, Down, Left action. ...

Similar publications

Figure 1. The schematic diagram of a proposed multi-degree optical...
Figure 6. Markov Decision Process in PCN environment. S T and S T+1 are...
The schematic diagram of a proposed multi-degree optical switch based...
The architecture of our proposed four-degree 3 × 3 × 3 × 3 silicon...
+12 Operation of the TOPS using the metallic Ti-heater (a) schematic...
Self-controlling photonic-on-chip networks with deep reinforcement learning
Article
Full-text available
  • Nov 2021
We present a novel photonic chip design for high bandwidth four-degree optical switches that support high-dimensional switching mechanisms with low insertion loss and low crosstalk in a low power consumption level and a short switching time. Such four-degree photonic chips can be used to build an integrated full-grid Photonic-on-Chip Network (PCN)....
View