On Learning Ising Models under Huber's Contamination Model

Authors: Adarsh Prasad, Vishwak Srinivasan, Sivaraman Balakrishnan, Pradeep Ravikumar

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We corroborate our theoretical results by simulations.
Researcher Affiliation Academia Machine Learning Department Department of Statistics and Data Science Carnegie Mellon University Pittsburgh, PA 15213
Pseudocode Yes Algorithm 1 Robust1DMean Robust univariate mean estimator
Open Source Code No The paper does not contain an explicit statement about releasing source code or a link to a code repository for the methodology described.
Open Datasets No The paper describes synthetic experiments where graphs are constructed with varying parameters, but it does not use or provide access information for a publicly available or open dataset.
Dataset Splits No The paper describes synthetic experiments and simulations, but it does not specify train, validation, and test dataset splits as typically done for machine learning experiments.
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU/CPU models, memory) used to run the experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers (e.g., library names, solvers, frameworks).
Experiment Setup Yes We generate our plots in the following manner: first we construct two graphs with the same structure either from Gclique p,d of Gstar p,d . We instantiate parameters for the first graph with θ(1) with model width ω and then vary the parameters for the second graph as θ(2) = θ(1) i 25 for i ranging from 1 to 50. We vary p {12, 15}, d {3 : 8 : 1} and ω {0.2 : 1.0 : 0.2} {1.5 : 10 : 0.5} where {a : b : c} denotes values between a and b (both inclusive) with consecutive values differing by c.