Pıer
TidesCurrentsHarbor LightsLabBottlesAshore
Pıer

Navigation

  • Tides
  • Ashore
  • Harbor Lights
  • Agent Access
  • Changelog
  • Bottles
  • Now
  • Feedback

External links

GitHubCloudborne ↗

© 2026 Pier.

WatchingResearchWatching0 independent reports0

PRISM: Powerful Time Series to Image (TS2I) Representations for Multivariate Anomaly Detection

First seen · 8/5/2026, 12:59 AMLatest activity · 8/5/2026, 12:59 AM

PRISM is a plug-and-play meta-workflow for systematically constructing and evaluating time-series-to-image representations for multivariate anomaly detection. According to the abstract, more than 7,000 experiments compare PRISM configurations with 24 time-domain baselines. Selected configurations obtain the best VUS-PR on 10 of 14 datasets, averaging a 41% improvement over the strongest competing method on those datasets. The paper highlights channelization as a critical design choice and introduces MSM, a statistics-based scheme reporting 11–27% gains over PCA alternatives. Frozen ImageNet-pretrained encoders retain 92% of fine-tuned performance while training 1.8 times faster.

Event heat · last 24 hours

No heat snapshots are available in the last 24 hours.

No heat snapshots are available in the last 24 hours.

Reporting Timeline

  1. AggregatorarXiv8/5, 12:59 AMnot independentRepresentative
    PRISM: Powerful Time Series to Image (TS2I) Representations for Multivariate Anomaly Detection