# S-TRIGGER: Continual State Representation Learning via Self-Triggered Generative Replay

arXiv: https://arxiv.org/abs/1902.09434 · HTML: https://mlsanity.com/paper/1902.09434/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Feb 2019.

## Built on by (most cited first, 7 of 7)

- [Replay in Deep Learning: Current Approaches and Missing Biological Elements](https://mlsanity.com/paper/2104.04132/) (2021)
- [A Survey of Continual Reinforcement Learning](https://mlsanity.com/paper/2506.21872/) (2025)
- [Self-Activating Neural Ensembles for Continual Reinforcement Learning](https://mlsanity.com/paper/2301.00141/) (2022)
- [Adiabatic replay for continual learning](https://mlsanity.com/paper/2303.13157/) (2023)
- [An Online Data-Driven Emergency-Response Method for Autonomous Agents in Unforeseen Situations](https://mlsanity.com/paper/2112.09670/) (2021)
- [Neural Distillation as a State Representation Bottleneck in Reinforcement Learning](https://mlsanity.com/paper/2210.02224/) (2022)
- [Investigation of D-Wave quantum annealing for training Restricted Boltzmann Machines and mitigating catastrophic forgetting](https://mlsanity.com/paper/2508.15697/) (2025)

## Abstract (arXiv)

We consider the problem of building a state representation model for control, in a continual learning setting. As the environment changes, the aim is to efficiently compress the sensory state's information without losing past knowledge, and then use Reinforcement Learning on the resulting features for efficient policy learning. To this end, we propose S-TRIGGER, a general method for Continual State Representation Learning applicable to Variational Auto-Encoders and its many variants. The method is based on Generative Replay, i.e. the use of generated samples to maintain past knowledge. It comes along with a statistically sound method for environment change detection, which self-triggers the Generative Replay. Our experiments on VAEs show that S-TRIGGER learns state representations that allows fast and high-performing Reinforcement Learning, while avoiding catastrophic forgetting. The resulting system is capable of autonomously learning new information without using past data and with a bounded system size. Code for our experiments is attached in Appendix.