Pith. sign in

REVIEW 3 cited by

Omni-Scale CNNs: a simple and effective kernel size configuration for time series classification

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2002.10061 v3 pith:FSSPGEJQ submitted 2020-02-24 cs.LG stat.ML

classification cs.LGstat.ML
keywords sizeos-blockseriestimekernelmodelsoptimalperformance
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The Receptive Field (RF) size has been one of the most important factors for One Dimensional Convolutional Neural Networks (1D-CNNs) on time series classification tasks. Large efforts have been taken to choose the appropriate size because it has a huge influence on the performance and differs significantly for each dataset. In this paper, we propose an Omni-Scale block (OS-block) for 1D-CNNs, where the kernel sizes are decided by a simple and universal rule. Particularly, it is a set of kernel sizes that can efficiently cover the best RF size across different datasets via consisting of multiple prime numbers according to the length of the time series. The experiment result shows that models with the OS-block can achieve a similar performance as models with the searched optimal RF size and due to the strong optimal RF size capture ability, simple 1D-CNN models with OS-block achieves the state-of-the-art performance on four time series benchmarks, including both univariate and multivariate data from multiple domains. Comprehensive analysis and discussions shed light on why the OS-block can capture optimal RF sizes across different datasets. Code available [https://github.com/Wensi-Tang/OS-CNN]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Amortized Predictability-aware Training Framework for Time Series Forecasting and Classification

    cs.LG 2026-02 conditional novelty 6.0 of 10

    APTF reweights training samples by loss-based predictability buckets and uses an amortization model to stabilize the estimates, improving accuracy across TSF and TSC baselines.

  2. QualityFM: a Multimodal Physiological Signal Foundation Model with Self-Distillation for Signal Quality Challenges in Critically Ill Patients

    cs.LG 2025-09 conditional novelty 5.0 of 10

    QualityFM, a multimodal ECG/PPG foundation model using self-distillation from clean to noisy signals, outperforms task-specific baselines on three ICU monitoring tasks.

  3. Finding Optimal Kernel Size and Dimension in Convolutional Neural Networks An Architecture Optimization Approach

    cs.CV 2025-06 reject novelty 3.0 of 10

    A proposed BKSEF heuristic for layer-wise CNN kernel sizes is presented, but the formula is ad hoc and the reported validation is missing from the paper.

Pith tools