Grasping Core Rules of Time Series through Pure Models

by   Gedi Liu, et al.

Time series underwent the transition from statistics to deep learning, as did many other machine learning fields. Although it appears that the accuracy has been increasing as the model is updated in a number of publicly available datasets, it typically only increases the scale by several times in exchange for a slight difference in accuracy. Through this experiment, we point out a different line of thinking, time series, especially long-term forecasting, may differ from other fields. It is not necessary to use extensive and complex models to grasp all aspects of time series, but to use pure models to grasp the core rules of time series changes. With this simple but effective idea, we created PureTS, a network with three pure linear layers that achieved state-of-the-art in 80 nearly the lightest model and having the fastest running speed. On this basis, we discuss the potential of pure linear layers in both phenomena and essence. The ability to understand the core law contributes to the high precision of long-distance prediction, and reasonable fluctuation prevents it from distorting the curve in multi-step prediction like mainstream deep learning models, which is summarized as a pure linear neural network that avoids over-fluctuating. Finally, we suggest the fundamental design standards for lightweight long-step time series tasks: input and output should try to have the same dimension, and the structure avoids fragmentation and complex operations.


page 1

page 2

page 3

page 4


Time Series Forecasting With Deep Learning: A Survey

Numerous deep learning architectures have been developed to accommodate ...

TimeGNN: Temporal Dynamic Graph Learning for Time Series Forecasting

Time series forecasting lies at the core of important real-world applica...

Time Series Forecasting with Ensembled Stochastic Differential Equations Driven by Lévy Noise

With the fast development of modern deep learning techniques, the study ...

Grasp stability prediction with time series data based on STFT and LSTM

With an increasing demand for robots, robotic grasping will has a more i...

Deep Learning in Multiple Multistep Time Series Prediction

The project aims to research on combining deep learning specifically Lon...

Dynamical Wasserstein Barycenters for Time-series Modeling

Many time series can be modeled as a sequence of segments representing h...

Predicting the Stability of Hierarchical Triple Systems with Convolutional Neural Networks

Understanding the long-term evolution of hierarchical triple systems is ...

Please sign up or login with your details

Forgot password? Click here to reset