Temporal Difference Learning as Gradient Splitting

Authors: Rui Liu, Alex Olshevsky

ICML 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Our main contribution is to provide an interpretation of temporal difference learning: we show how to view it as a splitting (a term we will define later) of an appropriately chosen quadratic form. As a consequence of this interpretation, it is possible to apply convergence proofs for gradient descent almost verbatim to temporal difference learning.The convergence times bounds we obtain this way improve on existing results.
Researcher Affiliation Academia 1Division of Systems Engineering, Boston University, Boston, MA, USA 2Department of ECE and Division of Systems Engineering, Boston University, Boston, MA, USA.
Pseudocode Yes Algorithm 1 Mean-adjusted TD(0)
Open Source Code No The paper does not provide any links or explicit statements about the availability of open-source code for the described methodology.
Open Datasets No The paper is theoretical and does not involve the use of specific datasets for training or experimentation. It discusses theoretical properties of Markov chains.
Dataset Splits No The paper is theoretical and does not describe training, validation, or test dataset splits.
Hardware Specification No The paper is theoretical and does not mention any specific hardware used for experiments.
Software Dependencies No The paper is theoretical and does not specify any software dependencies with version numbers.
Experiment Setup No The paper is theoretical and does not include details about an experimental setup, hyperparameters, or training configurations.