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; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1]
Online Optimal Tracking of Linear Systems with Adversarial Disturbances
Authors: Farnaz Adib Yaghmaie, Hamidreza Modares
TMLR 2023 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Simulation results compare the presented approach against the H control as well as the LQR control to show its superiority. In this section, we give our simulation results. |
| Researcher Affiliation | Academia | Farnaz Adib Yaghmaie EMAIL Faculty of Electrical Engineering Linköping University Linköping, Sweden Hamidreza Modares EMAIL The Department of Mechanical Engineering Michigan State University Michigan, USA |
| Pseudocode | Yes | Algorithm 1 summarizes the online state tracking procedure. ... Algorithm 1 Online state tracking algorithm |
| Open Source Code | No | The paper does not provide any explicit statement about releasing source code or a link to a code repository for the methodology described. |
| Open Datasets | No | The paper describes a dynamical system and reference signal with specific matrices and parameters (e.g., 'xk+1 = 1 1 0 1 xk + 1 0 0 1 uk + wk'). It also details the generation of various disturbances. This indicates the use of simulated or synthetic data rather than publicly available datasets. |
| Dataset Splits | No | The paper uses a simulated environment and does not rely on external datasets that would require explicit training, validation, or test splits. It mentions simulation runs for a duration T=10000 steps and evaluation for Teval=30 steps, which are simulation durations, not dataset splits. |
| Hardware Specification | No | The paper describes the simulation setup and use of a 'Gaussian noise generator (numpy.random.normal) by Numpy in Python', but does not specify any hardware details like CPU, GPU models, or memory used for running the simulations. |
| Software Dependencies | No | The paper mentions the use of 'Gaussian noise generator (numpy.random.normal) by Numpy in Python'. However, it does not provide specific version numbers for Python, Numpy, or any other software libraries, which are required for a reproducible description of software dependencies. |
| Experiment Setup | Yes | We select K in equation 12 as K = (R + BT Pr B)-1BT Pr A. We keep K unchanged during running the algorithm. We set H = 5, mr = 5, mw = 5, η = 0.0001 and initialize M = 0, P = 0 . We do not use any information about the dynamics of the reference signal; we only use measured outputs of the reference signal rk in this algorithm. We also do not use any information about the disturbance in this algorithm. ... We consider a quadratic cost with Q = 20I2, R = I2; that is ck = e T k Qek + u T k Ruk. |