Adaptive Skip Intervals: Temporal Abstraction for Recurrent Dynamical Models

Alexander Neitz, Giambattista Parascandolo, Stefan Bauer, Bernhard Schölkopf
8/14/2018
cs.LGstat.ML

Official code: neitzal/adaptive-skip-intervals (24 stars, last commit 2018-09-07).

Abstract

We introduce a method which enables a recurrent dynamics model to be temporally abstract. Our approach, which we call Adaptive Skip Intervals (ASI), is based on the observation that in many sequential prediction tasks, the exact time at which events occur is irrelevant to the underlying objective. Moreover, in many situations, there exist prediction intervals which result in particularly easy-to-predict transitions. We show that there are prediction tasks for which we gain both computational efficiency and prediction accuracy by allowing the model to make predictions at a sampling rate which it can choose itself.

View on arXivView PDF

Code Implementations(1)

neitzal/adaptive-skip-intervalsOfficial100%

Implementation of the paper "Adaptive Skip Intervals: Temporal Abstraction for Recurrent Dynamical Models"

242PythonAug 13, 20188 years ago

Similar Papers

USTEP: Spatio-Temporal Predictive Learning under A Unified View

83% match2023

Time-Aware World Model for Adaptive Prediction and Control

83% match2 repos152025

Multi-Rate Mixture of Experts for Accelerating Liquid Neural Network Training

82% match2026

OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning

82% match2 repos1.2k2023

Koopa: Learning Non-stationary Time Series Dynamics with Koopman Predictors

82% match1 repo2412023

Long-Horizon Textual World Modeling through Structured Reasoning

82% match2026

Cite this paper

@article{neitz2018adaptive,
  title  = {Adaptive Skip Intervals: Temporal Abstraction for Recurrent Dynamical Models},
  author = {Alexander Neitz and Giambattista Parascandolo and Stefan Bauer and Bernhard Schölkopf},
  year   = {2018},
  eprint = {1808.04768},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1808.04768}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1808.04768.svg)](https://codeofpaper.com/paper/1808.04768)

Discussion