ESTransformer | Nhi N. Truong

At the VNU-HCMUT AI Lab, I developed ESTransformer, a hybrid forecasting model that combines Holt-Winter’s exponential smoothing with transformer neural networks.

Key contributions:

Combined Transformer neural networks with Holt-Winters statistical methods, reducing prediction error by 15% and improving training efficiency by 60% on complex seasonal datasets
Led the full 9-month development cycle from problem analysis to publication as first author
Published at the Time Series for Health workshop at ICLR 2024 (Truong et al., 2024)

The model was evaluated on the M4 competition dataset containing 100,000 time series across six different frequencies (hourly to yearly), demonstrating comparable or improved performance compared to the state-of-the-art ESRNN model.

References

2024

ICLR WS

Hybrid Transformer and Holt-Winter’s Method for Time Series Forecasting

Nhi N. Truong, Duc Q. Nguyen, Jeffrey Gropp, and 1 more author

In Time Series for Health Workshop, International Conference on Learning Representations (ICLR), Mar 2024

Abs HTML PDF

Time series forecasting is an important research topic in machine learning due to its prevalence in social and scientific applications. Multi-model forecasting paradigm, including model hybridization and model combination, is shown to be more effective than single-model forecasting in the M4 competition. In this study, we hybridize exponential smoothing with transformer architecture to capture both levels and seasonal patterns while exploiting the complex non-linear trend in time series data. We show that our model can capture complex trends and seasonal patterns with moderately improvement in comparison to the state-of-the-arts result from the M4 competition.