Christoph Sawade

and perhaps `value(l::LogisticLoss, f::Vector)`, which returns the losses for all possible classes.

Oh! You guys are very quick. I am trying to order my thoughts: **Loss types** > The actually estimation algorithm would be different for these two types of losses. For...

What do you guys think of a bottom-up approach. Where we just start to try this on an example in a WIP branch and look how it feels? I added...

@lindahua: yes, I agree. A soon as I have time, I will start with the multi-class logreg. Currently, I just want to see how conceptual things are working. Feel free...

Regarding the `RegressionSolver` in principle, I also agree. However, I would like to build up on `Optim.jl` whenever it is possible in order to avoid double work.

Hey @jumutc. I agree that the API is probably not perfect and has to be adapted. However, I am a fan of starting with something simple and make it more...

My intuition is that everything that can be expressed as a reg. empir. risk can / should be included. So I like the idea to add also non-regression methods. However,...

Hm. Perhaps I am convinced that it makes sense to have such a distinction. @lendle, @lindahua: What do think about [this](https://github.com/JuliaStats/RegERMs.jl/commit/561ed87df9d60e01980d8d25efadc3cf55d42f0c)?

Agree, currently I don't see an application of `GenerativeLoss`, since the loss always compare a prediction with a ground-truth. Can I close that issue and discuss it again, when we...

Great Initiative! I agree with this abstraction. But your argumentation holds in general for all regularized empirical risk minimization approaches. Is it necessary to restrict the base package `RegressionBase.jl` to...