POLA: Online Time Series Prediction by Adaptive Learning Rates
arXiv:2102.08907
Abstract
Online prediction for streaming time series data has practical use for many real-world applications where downstream decisions depend on accurate forecasts for the future. Deployment in dynamic environments requires models to adapt quickly to changing data distributions without overfitting. We propose POLA (Predicting Online by Learning rate Adaptation) to automatically regulate the learning rate of recurrent neural network models to adapt to changing time series patterns across time. POLA meta-learns the learning rate of the stochastic gradient descent (SGD) algorithm by assimilating the prequential or interleaved-test-then-train evaluation scheme for online prediction. We evaluate POLA on two real-world datasets across three commonly-used recurrent neural network models. POLA demonstrates overall comparable or better predictive performance over other online prediction methods.
arXiv commentAccepted to ICASSP 2021
References in corpus (5)
- Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions
- No More Pesky Learning Rates
- Continual Unsupervised Representation Learning
- Online Learning for Time Series Prediction
- Efficient Online Hyperparameter Optimization for Kernel Ridge Regression with Applications to Traffic Time Series Prediction