PAC Learnability of Node Functions in Networked Dynamical Systems

Authors: Abhijin Adiga, Chris J Kuhlman, Madhav Marathe, S Ravi, Anil Vullikanti

ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We present experimental results using both synthetic and real-world networks to demonstrate how network structure and sample complexity influence the quality of the inferred system.
Researcher Affiliation Academia 1Biocomplexity Institute and Initiative, University of Virginia, Charlottesville, VA, USA. 2Also with the Department of Computer Science, University of Virginia, Charlottesville, VA, USA. 3Also with the Department of Computer Science, University at Albany SUNY, Albany, NY, USA.
Pseudocode Yes Algorithm 1 A consistent learner for positive examples.
Open Source Code No The paper does not provide any explicit statement or link for the release of source code for the described methodology.
Open Datasets No Table 1. Mined and synthetic networks, and their attributes. The paper refers to 'Jazz', 'NRV eu Emall' as mined networks and 'synthetic networks' but does not provide concrete access information (e.g., specific links, DOIs, or formal citations with author names and years) for these datasets.
Dataset Splits No The paper does not provide specific details regarding training, validation, and test dataset splits, such as percentages, absolute sample counts, or citations to predefined splits.
Hardware Specification No The paper does not provide specific hardware details such as GPU/CPU models, processor types, or memory amounts used for running the experiments. It only generally refers to 'running experiments'.
Software Dependencies No The paper does not list specific software dependencies with their version numbers (e.g., Python 3.8, PyTorch 1.9, CPLEX 12.4).
Experiment Setup Yes For each true assignment of thresholds to nodes, Algorithm 1 is used to estimate ten inferred threshold assignments. ... the number nt of configurations C (i.e., comparisons or trials) to be nt = 10n. ... different numbers m of queries (or configurations), ranging from 10 to 105. ... One configuration distribution Du is a uniform distribution where each node is set to state 1 with probability p (and to state 0 with probability (1 p)). ... For Du in the first plot, we take p = 0.25.