Nearly Optimal Robust Subspace Tracking
Authors: Praneeth Narayanamurthy, Namrata Vaswani
ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | 4. Empirical Evaluation The complete details for all experiments are provided in (Narayanamurthy & Vaswani, 2018b). All codes can be found at https://github.com/praneethmurthy/NORST.. Figure 1. Background Recovery. NORST is the only technique that works for both sequences. ... Figure 2. The plot of subspace error versus time... Table 2. Comparison of ˆL L F / L F for offline RPCA methods... |
| Researcher Affiliation | Academia | 1Department of Electrical and Computer Engineering, Iowa State University, USA. |
| Pseudocode | Yes | Algorithm 1 Basic-NORST (with tj known). |
| Open Source Code | Yes | All codes can be found at https://github.com/praneethmurthy/NORST.. |
| Open Datasets | No | The paper describes using 'Synthetic Data' and 'Video Frame' data, but does not provide concrete access information (link, DOI, formal citation, or name of an established public dataset) for either, nor does it explicitly state that the datasets are publicly available with access details. |
| Dataset Splits | No | The paper states 'We used the first ttrain = 3.3r = 100 frames as the training data with fewer outliers', but it does not specify explicit training, validation, and test splits (e.g., percentages or sample counts for each partition) or refer to a standard predefined split with citation. |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU/CPU models, memory, or cloud instance types used for running experiments. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers (e.g., 'Python 3.8', 'PyTorch 1.9'). |
| Experiment Setup | Yes | In 'Synthetic Data' section: 'In our experiments we use n = 1000, d = 12000, r = 30, J = 2, t1 = 3000, t2 = 8000, r = 30, δ1 = δ2 = 0.001. ... The algorithm parameters are set as K = 8, α = 300.' In 'Video' section: 'The NORST parameters were set as α = 60, K = 3, r = 40 and ξt = Ψˆℓt 1 .' |