Interpretable Graph-Based Semi-Supervised Learning via Flows

Authors: Raif Rustamov, James Klosowski

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We present an empirical evaluation both on synthetic and real data (Section 6).
Researcher Affiliation Industry Raif M. Rustamov, James T. Klosowski AT&T Labs Research
Pseudocode No The paper describes algorithms (e.g., ADMM in Section 5) but does not present them in a structured pseudocode block or algorithm figure.
Open Source Code No The paper does not provide any explicit statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes Next we test the proposed method on high-dimensional image and text datasets used in (Zhou and Belkin 2011), including MNIST 3vs8, MNIST 4vs9, aut-avn, ccat, gcat, pcmac, and real-sim.
Dataset Splits Yes In each run we use nl = 50 labeled samples; in addition, we withheld 50 samples for validation.
Hardware Specification No The paper does not provide specific details about the hardware used for experiments, such as CPU/GPU models, memory, or cloud instance types.
Software Dependencies No The paper mentions 'ADMM' and 'Graphviz' but does not specify version numbers for any software dependencies, libraries, or solvers used in the experiments.
Experiment Setup Yes For p-Voltages, the value of p for each run is chosen from {1.0625, 1.125, 1.25, 1.5, 2} using the best performing value on the validation samples... For iterated Laplacian method, m is chosen from {1, 2, 4, 8, 16}... For the flow approach, the value of λ is chosen from {0, 0.025, 0.05, 0.1, 0.2}...