Learning Reward Machines for Partially Observable Reinforcement Learning

NeurIPS 2019 • Rodrigo Toro Icarte • Ethan Waldie • Toryn Klassen • Rick Valenzano • Margarita Castro • Sheila Mcilraith

Reward Machines (RMs), originally proposed for specifying problems in Reinforcement Learning (RL), provide a structured, automata-based representation of a reward function that allows an agent to decompose problems into subproblems that can be efficiently learned using off-policy learning. Here we show that RMs can be learned from experience, instead of being specified by the user, and that the resulting problem decomposition can be used to effectively solve partially observable RL problems... (read more)

PDF Abstract

Results from the Paper

Add Remove

Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

METHOD	TYPE
Entropy Regularization	Regularization
Convolution	Convolutions
PPO	Policy Gradient Methods
Dense Connections	Feedforward Networks
Softmax	Output Functions
ReLU	Activation Functions
TRPO	Policy Gradient Methods
Retrace	Value Function Estimation
Experience Replay	Replay Memory
A3C	Policy Gradient Methods
Stochastic Dueling Network	Value Function Estimation
ACER	Policy Gradient Methods

METHOD

TYPE

Entropy Regularization