How could I check the convergence of uBoost when using uniforming_rate (alpha) != 0?. When I plot the log-loss metric vs number of boostings I see it increases, with a rate proportional to the alpha value used. You can see this trend in the plot attached. On the other hand, I can make the log-loss to converge with another hyper-parameter configuration (for the same alpha) but then I don't get an uniform selection. How can I deal with this?, does it mean that the log-loss is not a good metric to check convergence in this case?.
uboost_vs_adaboost.pdf
for uBoost convergence is something poorly defined.
first, uBoost has no optimization target (contrary say to AdaBoost, GBDT, GB+FL)
second, the way it operates quite often drives to bias in weights in one of direction - thus probabilities may become very biased, thus loss may diverge
Among options, I recommend to monitor ROC AUC on validation set or some similar discriminative measure.
Hello,
How could I check the convergence of uBoost when using uniforming_rate (alpha) != 0?. When I plot the log-loss metric vs number of boostings I see it increases, with a rate proportional to the alpha value used. You can see this trend in the plot attached. On the other hand, I can make the log-loss to converge with another hyper-parameter configuration (for the same alpha) but then I don't get an uniform selection. How can I deal with this?, does it mean that the log-loss is not a good metric to check convergence in this case?. uboost_vs_adaboost.pdf
Thanks very much, Gino