Adversarial Classification: Necessary Conditions and Geometric Flows

Authors: Nicolás García Trillos, Ryan Murray

JMLR 2022 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical examples illustrating these ideas are also presented.
Researcher Affiliation Academia Nicol as Garc ıa Trillos EMAIL Department of Statistics University of Wisconsin Madison, Wisconsin, USA Ryan Murray EMAIL Department of Mathematics North Carolina State University Raleigh, NC, USA
Pseudocode No The paper describes mathematical derivations and propositions but does not include any explicit pseudocode or algorithm blocks. For example, it mentions an "MBO scheme" as a numerical algorithm from prior work, but does not present its own structured pseudocode.
Open Source Code No The paper does not provide any concrete access information for source code, such as repository links or explicit statements of code release. It mentions using a "standard ODE solver in Python" and a "modified version of the scheme from (Merriman et al., 1992)" but these refer to external tools/methods rather than the authors' own implementation code.
Open Datasets No The paper uses synthetic data for its numerical examples, describing the probability distributions. For instance, in Section 4.1 it states: "Suppose that P(X dx|Y = 1) = φ(x)dx... and let P(X dx|Y = 0) = φ((x 2)/2)dx 2 ." In Section 6.3, it says: "We consider two different classes ρ1 N(( .5, 2), Σ) + N(( .5, .5), Σ) and ρ2 N((.5, .5), Σ) + N((.5, 2), Σ, where Σ = .2I, and w0 = w1 = .5." These are descriptions of data generation, not references to publicly available datasets.
Dataset Splits No The paper utilizes synthetic data generated from described distributions for its numerical examples. Therefore, the concept of specific training, validation, or test dataset splits is not applicable or mentioned.
Hardware Specification No The paper mentions running numerical examples using a "standard ODE solver in Python" and a "modified version of the scheme from (Merriman et al., 1992)" but does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used for these computations.
Software Dependencies No The paper mentions using a "standard ODE solver in Python" and a "modified version of the scheme from (Merriman et al., 1992)" for its numerical examples. However, it does not provide specific version numbers for Python or any of the solvers/schemes used, which is required for a reproducible description of ancillary software.
Experiment Setup No The paper describes the data distributions for its numerical examples (e.g., "Suppose that P(X dx|Y = 1) = φ(x)dx..." in Section 4.1, and "We consider two different classes ρ1 N(( .5, 2), Σ) + N(( .5, .5), Σ) and ρ2 N((.5, .5), Σ) + N((.5, 2), Σ, where Σ = .2I, and w0 = w1 = .5." in Section 6.3). However, it does not provide specific hyperparameters, training configurations, or system-level settings typically found in experimental setups for model training or evaluation.