Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..

Consistency of the $k_n$-nearest neighbor rule under adaptive sampling

Authors: Robi Bhattacharjee, Geelon So, Sanjoy Dasgupta

NeurIPS 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We study the kn-nearest neighbor learner within this setting. In the worst-case, the learner will fail because an adaptive process can generate spurious patterns out of noise. However, under the mild smoothing assumption that the process generating the instances is uniformly absolutely continuous and that choice of (kn)n is reasonable, the kn-nearest neighbor rule is online consistent.
Researcher Affiliation Academia Robi Bhattacharjee1 Sanjoy Dasgupta2 Geelon So2 1University of Tübingen and Tübingen AI Center 2Department of Computer Science and Engineering, UC San Diego
Pseudocode Yes Algorithm 1 The kn-nearest neighbor rule 1: for n = 1, 2, . . . do 2: Receive the instance Xn 3: Predict the majority vote label of the kn nearest neighbors X(1) n , . . . , X(kn) n , k=1 Y (k) n 1/2 4: Observe and memorize the label Yn 5: end for
Open Source Code No No experiments. (from checklist)
Open Datasets No No experiments. (from checklist)
Dataset Splits No No experiments. (from checklist)
Hardware Specification No No experiments. (from checklist)
Software Dependencies No No experiments. (from checklist)
Experiment Setup No No experiments. (from checklist)