# STCN: Stochastic Temporal Convolutional Networks

arXiv: https://arxiv.org/abs/1902.06568 · HTML: https://mlsanity.com/paper/1902.06568/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Feb 2019.

## Built on by (most cited first, 26 of 26)

- [Self-Supervised Speech Representation Learning: A Review](https://mlsanity.com/paper/2205.10643/) (2022)
- [Speech Enhancement and Dereverberation with Diffusion-based Generative Models](https://mlsanity.com/paper/2208.05830/) (2022)
- [AirFormer: Predicting Nationwide Air Quality in China with Transformers](https://mlsanity.com/paper/2211.15979/) (2022)
- [Handwriting Transformers](https://mlsanity.com/paper/2104.03964/) (2021)
- [Handwritten Text Generation from Visual Archetypes](https://mlsanity.com/paper/2303.15269/) (2023)
- [Unsupervised Speech Enhancement using Dynamical Variational Auto-Encoders](https://mlsanity.com/paper/2106.12271/) (2021)
- [Graph Neural Processes for Spatio-Temporal Extrapolation](https://mlsanity.com/paper/2305.18719/) (2023)
- [VATr++: Choose Your Words Wisely for Handwritten Text Generation](https://mlsanity.com/paper/2402.10798/) (2024)
- [SmartPatch: Improving Handwritten Word Imitation with Patch Discriminators](https://mlsanity.com/paper/2105.10528/) (2021)
- [A spatio-temporal network for video semantic segmentation in surgical videos](https://mlsanity.com/paper/2306.11052/) (2023)
- [Diverse Generation from a Single Video Made Possible](https://mlsanity.com/paper/2109.08591/) (2021)
- [Zero-Shot Styled Text Image Generation, but Make It Autoregressive](https://mlsanity.com/paper/2503.17074/) (2025)
- [A Brief Overview of Unsupervised Neural Speech Representation Learning](https://mlsanity.com/paper/2203.01829/) (2022)
- [VDSM: Unsupervised Video Disentanglement with State-Space Modeling and Deep Mixtures of Experts](https://mlsanity.com/paper/2103.07292/) (2021)
- [Time-Transformer: Integrating Local and Global Features for Better Time Series Generation (Extended Version)](https://mlsanity.com/paper/2312.11714/) (2023)
- [Ask "Who", Not "What": Bitcoin Volatility Forecasting with Twitter Data](https://mlsanity.com/paper/2110.14317/) (2021)
- [Autoregressive Styled Text Image Generation, but Make it Reliable](https://mlsanity.com/paper/2510.23240/) (2025)
- [TLNets: Transformation Learning Networks for long-range time-series prediction](https://mlsanity.com/paper/2305.15770/) (2023)
- [Benchmarking Generative Latent Variable Models for Speech](https://mlsanity.com/paper/2202.12707/) (2022)
- [Stochastic Diffusion: A Diffusion Probabilistic Model for Stochastic Time Series Forecasting](https://mlsanity.com/paper/2406.02827/) (2024)
- [FW-GAN: Frequency-Driven Handwriting Synthesis with Wave-Modulated MLP Generator](https://mlsanity.com/paper/2508.21040/) (2025)
- [Learning deep autoregressive models for hierarchical data](https://mlsanity.com/paper/2104.13853/) (2021)
- [LiP-Flow: Learning Inference-time Priors for Codec Avatars via Normalizing Flows in Latent Space](https://mlsanity.com/paper/2203.07881/) (2022)
- [Diverse Video Generation from a Single Video](https://mlsanity.com/paper/2205.05725/) (2022)
- [ScriptViT: Vision Transformer-Based Personalized Handwriting Generation](https://mlsanity.com/paper/2511.18307/) (2025)
- [SpiS-GAN: Spiral-Modulated Handwriting Synthesis with Star Operation](https://mlsanity.com/paper/2607.06949/) (2026)

## Abstract (arXiv)

Convolutional architectures have recently been shown to be competitive on many sequence modelling tasks when compared to the de-facto standard of recurrent neural networks (RNNs), while providing computational and modeling advantages due to inherent parallelism. However, currently there remains a performance gap to more expressive stochastic RNN variants, especially those with several layers of dependent random variables. In this work, we propose stochastic temporal convolutional networks (STCNs), a novel architecture that combines the computational advantages of temporal convolutional networks (TCN) with the representational power and robustness of stochastic latent spaces. In particular, we propose a hierarchy of stochastic latent variables that captures temporal dependencies at different time-scales. The architecture is modular and flexible due to the decoupling of the deterministic and stochastic layers. We show that the proposed architecture achieves state of the art log-likelihoods across several tasks. Finally, the model is capable of predicting high-quality synthetic samples over a long-range temporal horizon in modeling of handwritten text.