PulseAugur
EN
LIVE 15:23:36

Autoencoder models reduce runner telemetry to performance scores

This paper explores the use of autoencoder architectures for reducing complex wearable telemetry data from runners into a single performance score. Researchers evaluated five dimensionality reduction models, including three autoencoder variants and PCA, assessing them on reconstruction error and the interpretability of the resulting latent score. The study found that deep autoencoders performed best in terms of both low reconstruction error and high composite interpretability scores, with running pace, aerobic decoupling, and average heart rate identified as key drivers of the latent score. AI

IMPACT This research could lead to more sophisticated performance analysis tools for athletes using wearable technology.

RANK_REASON The cluster contains a research paper detailing novel methodology for data analysis.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

Autoencoder models reduce runner telemetry to performance scores

COVERAGE [2]

  1. arXiv cs.LG TIER_1 English(EN) · Mateusz Kubita, Jan Zubalewicz, Krzysztof Siwek ·

    Autoencoder Architectures for Athlete Performance Scoring from Wearable Telemetry

    arXiv:2606.28145v1 Announce Type: new Abstract: Wearable devices produce large, high dimensional training logs for everyday runners, and interpretation rather than data collection is now the limiting step. This paper evaluates five dimensionality reduction models, three autoencod…

  2. arXiv cs.LG TIER_1 English(EN) · Krzysztof Siwek ·

    Autoencoder Architectures for Athlete Performance Scoring from Wearable Telemetry

    Wearable devices produce large, high dimensional training logs for everyday runners, and interpretation rather than data collection is now the limiting step. This paper evaluates five dimensionality reduction models, three autoencoder variants, PCA, and a Variational Autoencoder,…