Tom Breloff
Tom Breloff
I don't remember if we distinguish actions from action_space, but that certainly makes sense. On Friday, September 16, 2016, Spencer Lyon [email protected] wrote: > Note that the action space is...
I'm going to make the following API changes: ``` julia actions(env) --> A done(env) --> bool # becomes: actions(env, s) --> A finished(env, s′) --> bool ``` and I'll change...
ref: f21b5cf
I want to link the solvers to the work I'm doing in StochasticOptimization, so there's still a lot of changes to be made here, but hopefully one could develop the...
I would say that I haven't settled on a policy api yet... I've been a little more focused on the environments. If you have time, could you write out a...
I think that, without sample code, I'll have a hard time understanding what a "getter/setter" is. Do you mean a lookup table for states and actions? If so, my interest...
So that's not really reinforcement learning. You should check out our effort in JuliaML if you're more interested in more general machine learning. In RL there are no "answers", only...
I would consider adding AdaMax as well... it's very briefly referenced in the Adam paper, but I think it's actually better.
@svs14 @Rory-Finnegan We've been having a very similar discussion over at https://github.com/Evizero/LearnBase.jl/issues/2... it would be great if you could weigh in and also let us know if you might envision...
Simon: can I make you collaborator on MetaPkg so you keep that up to date? The nice thing about recommending MetaPkg over a list of checkout commands is that you...