Instance Based Approximations to Profile Maximum Likelihood

Authors: Nima Anari, Moses Charikar, Kirankumar Shiragur, Aaron Sidford

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

Reproducibility Variable Result LLM Response
Research Type Experimental We provide preliminary experiments in which we perform entropy estimation using the Pseudo PML approach implemented using our simpler rounding algorithm. Our results match other state-of-the-art estimators for entropy, some of which are property specific.
Researcher Affiliation Academia Nima Anari Stanford University anari@stanford.edu Moses Charikar Stanford University moses@cs.stanford.edu Kirankumar Shiragur Stanford University shiragur@stanford.edu Aaron Sidford Stanford University sidford@stanford.edu
Pseudocode Yes Algorithm 1 Approximate PML(φ, R) and Algorithm 2 Approximate PML2(φ, R) are presented with clear, numbered steps.
Open Source Code No The paper mentions using external tools (CVX[GB14] with package CVXQUAD[FSP17]) for implementation, but it does not explicitly state that its own source code is open or provide any links to a repository for the methodology described.
Open Datasets No The paper refers to generating data from
Dataset Splits No The paper discusses 'sample size' in the context of experiments but does not provide specific details on train/validation/test dataset splits or cross-validation setups.
Hardware Specification No The paper does not provide any specific hardware details (e.g., CPU/GPU models, memory, or cloud instances) used for running the experiments.
Software Dependencies No In our implementation we use CVX[GB14] with package CVXQUAD[FSP17] to solve the convex program. However, specific version numbers for these software packages are not provided.
Experiment Setup No The paper describes algorithms and theoretical guarantees, but it does not provide specific experimental setup details such as hyperparameter values, optimizer settings, or training schedules.