Learning Nearest Neighbor Graphs from Noisy Distance Samples

Authors: Blake Mason, Ardhendu Tripathy, Robert Nowak

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

Reproducibility Variable Result LLM Response
Research Type Experimental In Section 5 we show ANNTri s empirical performance on both simulated and real data. In particular, we highlight its efficiency in learning from human judgments.
Researcher Affiliation Academia Blake Mason University of Wisconsin Madison, WI 53706 bmason3@wisc.edu Ardhendu Tripathy University of Wisconsin Madison, WI 53706 astripathy@wisc.edu Robert Nowak University of Wisconsin Madison, WI 53706 rdnowak@wisc.edu
Pseudocode Yes Algorithm 1 ANNTri and Algorithm 2 SETri are provided with structured steps.
Open Source Code Yes Implementations of ANNTri, ANN, and RANDOM can be found alongside a demo and summary slides at https://github.com/blakemas/nngraph.
Open Datasets Yes For this experiment, we used a set X of 85 images of shoes drawn from the UT Zappos50k dataset [32, 33]
Dataset Splits No The paper mentions 'cross validation' for selecting an embedding dimension, but does not provide specific training/test/validation dataset splits (percentages, counts, or predefined standard splits) for its experiments.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory amounts, or cluster specifications) used for running its experiments.
Software Dependencies No The paper mentions general tools and algorithms (e.g., 'Python array/Matlab notation', 'STE algorithm') but does not list specific software components with their version numbers.
Experiment Setup Yes To construct the tightest possible confidence bounds for SETri, we use the law of the iterated logarithm as in [18] with parameters = 0.7 and δ = 0.1. Our analysis bounds the number of queries made to the oracle. We visualize the performance by tracking the empirical error rate with the number of queries made per point. ... All rounds were capped at 105 samples for efficiency.