yuta-saito
yuta-saito
@Braamling Thank you for your contribution! and let me get back to you sometime next week.
@Braamling Thank you so much again for your contributions! I've checked your implementations and actually found them to be controversial in the following aspects. In my understanding, it seems that...
@Braamling Great to hear your plan for the workshop paper, and thank you for your suggestions! I agree that expanding the package by implementing an online simulation environment makes a...
@Braamling Thank you, and the high-level implementation looks great! (though I haven't yet gone through every detail) - Would you be able to provide the notebooks? Thot would be very...
@Braamling Sorry for not being able to respond, because I'm getting into the new semester. I've started reviewing the new content, and the notebook examples look great. Let me see...
@Braamling Thank you so much again for the contributions! I made some minor comments above. I'd also suggest renaming the notebook examples as follows. - [delay-simulation.ipynb](https://github.com/Braamling/zr-obp/blob/bvandenakker/online-simulator-with-drift-and-delays/examples/quickstart/delay-simulation.ipynb) -> online-bandit-with-delay-simulation.ipynb - [drift-simulation.ipynb](https://github.com/Braamling/zr-obp/blob/bvandenakker/online-simulator-with-drift-and-delays/examples/quickstart/drift-simulation.ipynb)...
@Braamling Thank you so much for the reminder, and I'm happy to merge, but it seems that some tests are failing, so it would be great if you could deal...
@arita37 Calibrating your pscore estimate might help, but depends on the situation. The following papers explore the effect of calibration in OPE, so might be of your interest: - Aniruddh...
@Braamling Looks great! Thank you so much again for this and also your contribution to our workshop at RecSys!
@Zziwei Thank you for your comments! > But after transforming the explicit data to implicit version, you do not filter out users who have 0 relevant items. In this case,...