5 skills found
incognite-lab / MyGymmyGym is the robotic simulator. It allows to create novel multi step long horizon tasks without coding. There are automatic task building routines, where user composes long horizon actions from basic set of protoactions, that are automatically paired with rewards. There are several humanoids and robotic arms ready to train.
Breakend / MultiStepBootstrappingInRLHere, we compare Q(\sigma) learning presented by Sutton and Barto in [1] to Tree-Backup, n-step Expected Sarsa, and n-step Sarsa.
JunjieYang97 / GGS MAML RLExample RL code for paper "Multi-Step Model-Agnostic Meta-Learning: Convergence and Improved Algorithms"
ropiens / Project Sandwich ManA project for researching a complex and long-horizon manipulation task especially focused on hierarchically stacking blocks.
manantomar / Multi Step Greedy Reinforcement Learning AlgorithmsMulti-step Greedy Reinforcement Learning Algorithms
