News for hackers. Stuff that matters.
Re:Impala: Scalable Distributed Deep Reinforcement Learning by Maybestring on Friday February 09, 2018 @12:24PM (#16342036)
If you allow untrusted users to define tasks for shared agents, then you must assume you will have malicous or incompetent users defining 'poison' tasks. These would be tasks that bias agents toward taking unproductive or harmful actions.

Afaik, this problem hasn't been solved.

Re:Impala: Scalable Distributed Deep Reinforcement Learning by radarsat1 on Friday February 09, 2018 @12:35PM (#16342130)
While you make a fair point, who mentioned anything about untrusted users? You seem to be off on a tangent..
Re:Impala: Scalable Distributed Deep Reinforcement Learning by Maybestring on Friday February 09, 2018 @12:49PM (#16342255)
The question is about running a node in a distributed learning algorithm. The algorithm depends on trusted nodes to function.
Re:Impala: Scalable Distributed Deep Reinforcement Learning by lawlessone on Friday February 09, 2018 @12:45PM (#16342211)
"untrusted users"

I mean using this at home on my own machine, not uploading it..

Re:Impala: Scalable Distributed Deep Reinforcement Learning by Maybestring on Friday February 09, 2018 @12:50PM (#16342265)
I don't understand the question then. You want to run an actor disconnected from the learner, but you want it to learn?

Edit: Or... You mean running the whole distributed learning algorithm on one machine. I don't see why not, but I doubt it would outperform something intended to run on a single machine.