Discrepancy in Tabula Rasa Returns: Are My Values Correct?

[sitaharsha]

I'm working on replicating the results of the Tabula Rasa experiment from the research paper. However, the values I obtained for maximum and average returns are different from those reported in the paper. I am using seed values 0, 1, 2, 3, and 4 for my calculations.

In the paper, the reported values are:

Maximum Returns: 1949.2 ± 247.3
Average Returns: 984.4 ± 26.9
When I run my code, I get the following values:

Overall Maximum env_loop/EpisodeReturn: 4822.2 ± 442.9
Overall Average env_loop/EpisodeReturn: 2784.8 ± 335.0

Can you please help me understand if these values are legitimate or if I made a mistake in my calculations? If there's an error in my code or approach, please guide me on how to fix it. I appreciate any assistance you can provide.

Thank you!

Additionally, I am ready to share my code if it helps in resolving this issue.

[sitaharsha]

Hi there,

I hope this message finds you well. I'm writing to follow up on an issue I raised a couple of weeks ago regarding a discrepancy I've found in the Tabula Rasa returns. I'm still interested in understanding this issue and resolving it if possible.

I've been attempting to further investigate the issue on my end, but I feel that some guidance would be immensely helpful. I'm also ready to share my code if it might help in pinpointing where the issue might lie.

Thank you for your time and I'm looking forward to your response.