Negotiable Reinforcement Learning for Pareto Optimal Sequential Decision-Making

Authors: Nishant Desai, Andrew Critch, Stuart J. Russell

NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In Section 4, we implement a simple NRL agent and make empirical observations of this bet settling behavior. Our experiments are run in a modified version of the Frozen Lake environment in Open AI Gym [Brockman et al., 2016].
Researcher Affiliation Academia Nishant Desai Center for Human-Compatible AI University of California, Berkeley nishantdesai@berkeley.edu Andrew Critch Department of EECS University of California, Berkeley critch@berkeley.edu Stuart Russell Computer Science Division University of California, Berkeley russell@cs.berkeley.edu
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any explicit statement about open-sourcing the code for the methodology or a link to a code repository.
Open Datasets Yes Our experiments are run in a modified version of the Frozen Lake environment in Open AI Gym [Brockman et al., 2016].
Dataset Splits No The paper describes a reinforcement learning setup using a modified Frozen Lake environment. It does not specify explicit training, validation, and test dataset splits as typically found in supervised learning contexts.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU, GPU models, or memory specifications) used for running the experiments.
Software Dependencies No The paper mentions 'Open AI Gym [Brockman et al., 2016]' and 'point-based value iteration [Pineau et al., 2003]' but does not provide specific version numbers for these or any other software dependencies.
Experiment Setup Yes After running point-based value iteration [Pineau et al., 2003] with a belief set of 331 points, we execute the resulting policy in this environment. The agent is initialized with initial belief state w1, corresponding to a subjective belief that the agent is in Principal 1 s MDP, M1, with probability w1 and Principal 2 s MDP, M2, with probability 1 w1 = w2.