Chapter 9 · Part 4
Deep Learning for Time Series
Modern deep architectures for forecasting — RNN/LSTM/GRU, the MLP family (N-BEATS, N-HiTS, DLinear, TSMixer), hybrid and modular designs, and emerging non-attention architectures including Mamba and diffusion-based forecasters.
Learning objectives
By the end of this chapter you will be able to:
- ▸Choose between MLP-based, RNN-based, and attention-based architectures
- ▸Train N-BEATS / N-HiTS / DLinear / TSMixer end-to-end
- ▸Recognise when a simple MLP beats a transformer on real benchmarks
- ▸Evaluate Mamba and diffusion-based forecasters honestly
- 1.1Introduction: From Classical to Deep Time Series Forecasting
- 2.1Recurrent Neural Networks
- 3.1Convolutional Neural Networks for Time Series Forecasting: Historical Development
- 4.1Multi-Layer Perceptrons for Time Series Forecasting: N-BEATS, N-HiTS, DLinear, TSMixer
- 5.1Hybrid and Modular Architectures
- 6.1Emerging Non-Attention Architectures: Mamba and Diffusion Models
- 7.1Performance Analysis Across Model Categories
- 8.1The Role of Hybrid Models in Forecasting
Get the full chapter
All eleven chapters · 478 references · private Discord community.
← Previous · Chapter 8
Advanced Machine Learning for Forecasting
Next · Chapter 10 →
Transformers for Time Series