We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Updated Loss functions (markdown)
Add explanation for loss function implementations.
as zak discovered, squared loss is not divided by 2 in loss_functions.cc
Fixed table formatting
Clarified minimizer for poisson regression
Added poisson loss (stub)
Added note on default loss function override for VWClassifier
Replaced missing video with the slides from same talk
Video not found.
make the png smaller
quantile loss formula png (images from sciweavers.org are not permanent URLs)
fixing the formula for quantile loss: tau means predicting tau-quantile, i.e. tau=0.1 results in lower predictions than tau=0.9
Quantile loss is median minimizer
fixed layout table
Added images for the loss functions (if someone can include in repo, thats nicer), added table, added link to John Langford's course. added info for "classic loss"
cleanup the links
add links to wikipedia