Reward-Free Kernel-Based Reinforcement Learning

Authors: Sattar Vakili, Farhang Nabiei, Da-Shan Shiu, Alberto Bernacchia

ICML 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We consider a broad range of kernel-based function approximations, including non-smooth kernels, and propose an algorithm based on adaptive domain partitioning. We show that our algorithm achieves order-optimal sample complexity for a large class of common kernels, which includes Mat ern and Neural Tangent kernels.
Researcher Affiliation Industry 1Media Tek Research. Correspondence to: Sattar Vakili <sattar.vakili@mtkresearch.com>.
Pseudocode Yes Algorithm 1 Exploration Phase, Algorithm 2 Planning Phase
Open Source Code No The paper does not provide any statement or link regarding the availability of open-source code for the described methodology.
Open Datasets No The paper is theoretical and focuses on sample complexity analysis for a reward-free RL framework, rather than conducting empirical studies on a specific dataset. Therefore, it does not mention dataset availability for training.
Dataset Splits No The paper is theoretical and does not involve empirical validation on specific datasets, hence no dataset split information is provided.
Hardware Specification No The paper is theoretical and focuses on algorithm design and sample complexity analysis. It does not mention any specific hardware used for experiments.
Software Dependencies No The paper is theoretical and describes algorithms and mathematical analysis. It does not mention any specific software dependencies or versions.
Experiment Setup No The paper is theoretical and focuses on algorithm design and sample complexity analysis. It does not describe any specific experimental setup, hyperparameters, or training configurations.