AI Paper Atlas

Research map / Time series / Foundation models for time series

A deep convolutional neural network that is invariant to time rescaling

Brandon G Jacques, Zoran Tiganj, Aakash Sarkar, Marc Howard, Per Sederberg

ICML 2022 · Time series › Foundation models for time series · approach: Time series representation learning · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use logarithmically compressed temporal memory with convolution and max pooling instead of a TCN's linear-time memorySITHCon vs. TCN · Morse Decoder, unseen time scalesaccuracyImproved
Use logarithmically compressed temporal memory with convolution and max pooling instead of a TCN's linear-time memorySITHCon vs. TCN · Morse Addition, unseen time scalesmean squared errorImproved
Use logarithmically compressed temporal memory with convolution and max pooling instead of a TCN's linear-time memorySITHCon vs. TCN · AudioMNIST, unseen time scalesaccuracyImproved
Add log-spaced temporal memory nodes to a trained SITHCon network (ablation)SITHCon · Morse Decoder, maximum time-rescaling with 100% accuracymaximum time-rescalingImproved
Train on multiple time scales rather than only scale 1.0SITHCon · Morse Decoder, intermediate and unseen scalesaccuracyImproved
Train on multiple time scales rather than only scale 1.0TCN · AudioMNIST, training scalesaccuracyHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy