Computation and Communication Efficient Adaptive Federated Optimization of Federated Learning for Internet of Things
Round 1
Reviewer 1 Report
Chen et al. present an enhanced version of Federated Learning, which effectively reduces learning errors by jointly considering two key variables: local update and parameter compression. However, there are a few issues that require further attention:
Figure 1 appears to be unclear, particularly the text description within the figure. It is recommended that the authors improve the resolution of the figure to enhance its readability.
Throughout the draft, there are several blank square annotations after specific lines, such as line 289 and line 279. It is advised that the authors remove these annotations to maintain the document's clarity.
Line 379 shows Algorithm 2, but no description or explanation of the Algorithm's source is provided in the draft. The authors may want to include relevant information about the Algorithm's origin for a more comprehensive understanding.
Moderate editing of English language required
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Dear Authors,
Please find my suggestions below.
Your manuscript is dealing with computation and communication efficient related to federated learning for internet of things. You did amazing job with this manuscript. Congratulations. There are few technical issues related to your manuscript. Please find my suggestions below.
Explain all abbreviations when you use them for the first time in your manuscript.
In the line 54 explain abbreviation.
Tell more about application of your novel efficient adaptive federated optimization in practice.
Wish you all the best in future work.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper proposes a novel efficient adaptive federated optimization (FedEAFO) algorithm to improve efficiency of FL, which minimizes the learning error via jointly considering two variables including local update and parameter compression. The topic is interesting, and the logic is complete. But some improvements can be conducted:
1. In practical applications of internet of Things, different IOT devices will be used and different kinds of data will be collected. How does Federated Learning to handle different types of data?
2. In Page 14, there is an algorithmic pseudo-code named “Algorithm 2”, but “Algorithm 2” is not mentioned in the main text.
3. The font size is not uniform, e.g., the caption of Figure 3.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
This revised version has successfully addressed all my concerns.
Author Response
Thank you very much for your suggestions.