Making steady progress, always.
Lead-dev of Anyscale's Ray RLlib. We're hiring!
-
anyscale.com
- Berkeley, CA; Düsseldorf, Germany
- https://linkedin.com/in/sven-mika
Highlights
- Pro
- 20 discussions answered
Block or Report
Block or report sven1977
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
-
ray-project/ray Public
An open source framework that provides a simple, universal API for building distributed applications. Ray is packaged with RLlib, a scalable reinforcement learning library, and Tune, a scalable hyp…
-
rlgraph/rlgraph Public
RLgraph: Modular computation graphs for deep reinforcement learning
-
-
tensorforce/tensorforce Public
Tensorforce: a TensorFlow library for applied reinforcement learning
-
-
924 contributions in the last year
Less
More
Contribution activity
February 2022
Created 3 commits in 1 repository
Created a pull request in ray-project/ray that received 3 comments
[RLlib] Speedup A3C up to 3x (new training_iteration function instead of execution_plan) and re-instate Pong learning test.
This PR:
Provides a new training_iteration function for A3C (alternative to existing execution_plan).
By default, uses that new iteration function (
+128
−43
•
3
comments
Opened 2 other pull requests in 1 repository
Reviewed 10 pull requests in 1 repository
ray-project/ray
10 pull requests
- [RLlib] Discussion 2022: Fix batch_mode="complete_episodes" documentation inaccuracy.
-
[RLlib] Issue 22036: Client should handle concurrent episodes with one being
training_enabled=False. - [RLlib] [CI] Deflake longer running RLlib learning tests for off policy algorithms. Fix seeding issue in TransformedAction Environments
- [RLlib] Cleanup SlateQ algo; add test + add target Q-net
-
[RLlib] Add
on_sub_environment_createdto DefaultCallbacks class. - [RLlib] AlphaStar: Parallelized, multi-agent/multi-GPU learning via league-based self-play.
-
[RLlib] Neural-MMO
keep_per_episode_custom_metricspatch (toward making Neuro-MMO RLlib's default massive-multi-agent learning test environment). -
[RLlib] Request CPU resources in
Trainer.default_resource_request()if using dataset input. - [RLlib] Add an env wrapper so RecSim works with our Bandits agent.
- [RLlib] Move bandit example scripts into examples folder.

