How’d you come up with the values for your grid search?
Andrew Olton
1

Some of these values are suggested in the original paper, which states that the default values are good to handle most problems. For the carrying capacity, I selected some numbers based on the expected number of users. Also, this is a non-exhaustive grid search, so by no means the optimal parameters should be contained within the grid. A random search or a Bayesian search may be more appropriate but this was out of the scope of the experiment.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.