---
title: "NN-kNN for Regression: Accurate Prediction from Interpretable Retrieval"
authors:
  - "Xiaomeng Ye"
  - "Yu Wang"
  - "David Leake"
  - "David Crandall"
  - "Great Abhieyighan"
  - "Mereck McGowan"
doi: "10.24963/ijcai.2026/93"
url: "https://ijcai.org/proceedings/2026/93"
pdf: "https://ijcai.org/proceedings/2026/0093.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "826-834"
track: "Main Track"
keywords:
  - "AI Ethics, Trust, Fairnes: Explainability and interpretability"
  - "Machine Learning: Regression"
  - "Knowledge Representation and Reasoning: Case-based reasoning"
  - "Machine Learning: Trustworthy machine learning"
  - "AI Ethics, Trust, Fairnes: Trustworthy AI"
---

# NN-kNN for Regression: Accurate Prediction from Interpretable Retrieval

Xiaomeng Ye, Yu Wang, David Leake, David Crandall, Great Abhieyighan, Mereck McGowan

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 826-834. DOI: [10.24963/ijcai.2026/93](https://doi.org/10.24963/ijcai.2026/93)

## Abstract

Neural Network k-Nearest Neighbor (NN-kNN) was proposed as an interpretable network model that learns feature weights and similarity to retrieve relevant cases for classification. This paper extends it to regression with the goal of generating accurate predictions based on neighboring cases with similar labels. Specifically, we introduce three modular components: an attention mechanism that weights the contribution of retrieved cases, a locality-aware regularizer that favors label-similar neighbors, and an optional case adaptation module that refines the retrieved estimate. Across synthetic and standard tabular regression benchmarks, NN-kNN achieves competitive predictive error against strong baselines (kNN-R, MLKR, and MLPs) while providing cases with similar labels as explanation (later referred as label-similar case). Moreover, NN-kNN supports manual knowledge injection through tuning weights for human-comprehensible features. The result is a simple, general, and interpretable approach to continuous-valued prediction that unifies retrieval, attention, and optional case adaptation within a single neural framework.