ICACT20240375 Question.1
Questioner: emailofzhoudi@163.com 2024-02-03 오후 3:21:49 |
ICACT20240375 Answer.1
Answer by Auhor 202212666@mail.sdu.edu.cn 2024-02-03 오후 3:21:49
Chrome  Click!! |
| What are the innovations in the algorithm that you mentioned in this paper? |
Thank you for the question. This article takes an improved version of the SAC algorithm by introducing an explorer module that allows the agent to maximize the exploration of unknown actions in order to prevent falling into a locally optimal outcome. |
ICACT20240375 Question.2
Questioner: xhxhn999@163.com 2024-02-05 오전 10:06:27 |
ICACT20240375 Answer.2
Answer by Auhor 202212666@mail.sdu.edu.cn 2024-02-05 오전 10:06:27
Chrome Click!! |
| What are the advantages of your proposed algorithm over traditional algorithms? |
Thank you for your question. Compared with the traditional SAC algorithm, the article introduces the explorer module, which can prevent the agent from falling into the local optimum, and has achieved a significant effect beamforming effect. |
ICACT20240375 Question.3
Questioner: madj0212@163.com 2024-02-05 오후 1:55:44 |
ICACT20240375 Answer.3
Answer by Auhor 202212666@mail.sdu.edu.cn 2024-02-05 오후 1:55:44
Chrome Click!! |
| Where are you going to look next, and is there a better algorithm? |
Thank you for your question. The next step is to start with multi-agent reinforcement learning, currently planning to go into the MADDPG algorithm. |
ICACT20240375 Question.4
Questioner: ssdmu@dlmu.edu.cn 2024-02-05 오후 2:24:05 |
ICACT20240375 Answer.4
Answer by Auhor 202212666@mail.sdu.edu.cn 2024-02-05 오후 2:24:05
Chrome Click!! |
| Normally it is an ideal RIS with a default phase of 1. How is this RIS phase loss designed? |
For the general case, the default RIS has a magnitude of 1, as we consider the actual scenario. There will be hardware loss, so we perform the beamforming design of RIS by introducing a hardware loss expression for RIS, which has been given as a source in the article. |
ICACT20240375 Question.5
Questioner: xhxhn999@163.com 2024-02-06 오전 9:46:15 |
ICACT20240375 Answer.5
Answer by Auhor 202212666@mail.sdu.edu.cn 2024-02-06 오전 9:46:15
Chrome Click!! |
| What are the advantages of a deep reinforcement learning-based approach over traditional approaches? |
For complex communication environments, traditional algorithms are pre-trained for specific environments, making it difficult to adapt to complex and changing communication environments, whereas deep reinforcement learning is able to interact with the environment in real time to adjust the latest optimization scheme. |
