Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Improved Regret Bounds for Thompson Sampling in Linear Quadratic Control Problems
Authors: Marc Abeille, Alessandro Lazaric
ICML 2018 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally, we report numerical simulations supporting the conjecture that our result extends to multi-dimensional systems. and Numerical simulations. Since all the results in Sect. 5 hold for n 1, we try to simulate several random LQ systems of variable dimensionality and numerically estimate the probability of being optimistic (popt) in each of them. |
| Researcher Affiliation | Industry | 1Criteo, Paris, France 2Facebook AI Research, Paris, France. |
| Pseudocode | Yes | Figure 1: Thompson sampling algorithm for LQR |
| Open Source Code | No | The paper does not provide any statement or link for the open-source code of the described methodology. |
| Open Datasets | No | The numerical simulations describe generating random LQ systems rather than using a pre-existing, publicly available dataset with concrete access information. |
| Dataset Splits | No | The paper does not describe specific dataset splits (e.g., training, validation, test percentages or counts) as it focuses on theoretical analysis and simulations of system parameters, not traditional dataset-based experiments. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for running the numerical simulations. |
| Software Dependencies | No | The paper does not specify any software dependencies with version numbers for reproducing the experiments. |
| Experiment Setup | Yes | We construct S by setting D = 20J(θ ). For different values of n and d, we sample θ as θ [i, j] N(0, 1) independently and we run multiple trajectories of TS of length T = 500 steps. At each step t, we sample 1000 eθt from the TS distribution (with rejection)... |