You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If I understood correctly, what you meant by low varentropy is high varentropy, and what you meant by high varentropy is low varentropy
For example, a low varentropy low entropy is a concentrated but evenly spreaded out distribution, e.g., p = [0.5, 0.5, 0, 0, ...]. It make sense to do branching here, though the quadrant says argmax.
In comparison, a high varentropy low entropy is a distribution with heavy long tail, e.g. p = [0.75, 0.001, 0.001, ...]. It seems like we should do argmax here, but the qaudrant plot says branching?
Can you check if you mixed up them?
The text was updated successfully, but these errors were encountered:
If I understood correctly, what you meant by low varentropy is high varentropy, and what you meant by high varentropy is low varentropy
For example, a low varentropy low entropy is a concentrated but evenly spreaded out distribution, e.g., p = [0.5, 0.5, 0, 0, ...]. It make sense to do branching here, though the quadrant says argmax.
In comparison, a high varentropy low entropy is a distribution with heavy long tail, e.g. p = [0.75, 0.001, 0.001, ...]. It seems like we should do argmax here, but the qaudrant plot says branching?
Can you check if you mixed up them?
The text was updated successfully, but these errors were encountered: