Making RL with Preference-based Feedback Efficient via Randomization

Authors: Runzhe Wu, Wen Sun

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

Reproducibility Variable Result LLM Response
Research Type Theoretical Overall, while our main contribution is on the theoretical side, our theoretical investigation provides several new practical insights.
Researcher Affiliation Academia Runzhe Wu Department of Computer Science Cornell University rw646@cornell.edu Wen Sun Department of Computer Science Cornell University ws455@cornell.edu
Pseudocode Yes Algorithm 1 Preference-based and Randomized Least-Squares Value Iteration (PR-LSVI)
Open Source Code No The paper does not provide any statement or link indicating the availability of open-source code for the described methodology.
Open Datasets No The paper describes theoretical algorithms and does not perform experiments on specific datasets. Therefore, no information about publicly available or open datasets for training is provided.
Dataset Splits No This is a theoretical paper and does not describe experiments with dataset splits. No specific dataset split information (percentages, sample counts, or citations to predefined splits) for training, validation, or testing is provided.
Hardware Specification No The paper is theoretical and does not describe the execution of experiments requiring specific hardware. Therefore, no hardware specifications (e.g., GPU/CPU models, memory details) are mentioned.
Software Dependencies No The paper is theoretical and focuses on algorithm design and analysis rather than practical implementation details. No specific software dependencies with version numbers (e.g., Python 3.8, PyTorch 1.9) are mentioned.
Experiment Setup No The paper is theoretical and defines algorithm parameters (e.g., sigma_r, sigma_P, epsilon) as part of the theoretical framework rather than specific experimental setup details for empirical evaluation. No section titled 'Experimental Setup' or similar detailing training configurations or system-level settings for experiments is present.