# Scene-Adaptive Continual Learning for CSI-based Human Activity Recognition with Mixture of Experts

arXiv: https://arxiv.org/abs/2605.06447 · HTML: https://mlsanity.com/paper/2605.06447/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Mixture of LoRA experts, one of 12 directions in Mixture-of-experts.
- Builds on 7 of its 13 references that are in the map.
- First posted on arXiv 7 May 2026.

## Placed in

- Mixture-of-experts / [Mixture of LoRA experts](https://mlsanity.com/moe/mixture-of-lora-experts/)

## Builds on (most cited first, 7 of 7)

- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [A Model or 603 Exemplars: Towards Memory-Efficient Class-Incremental Learning](https://mlsanity.com/paper/2205.13218/) (2022)
- [A Statistical Theory of Regularization-Based Continual Learning](https://mlsanity.com/paper/2406.06213/) (2024)
- [Transfer Learning in Human Activity Recognition: A Survey](https://mlsanity.com/paper/2401.10185/) (2024)
- [Optimal preprocessing of WiFi CSI for sensing applications](https://mlsanity.com/paper/2307.12126/) (2023)

## Abstract (arXiv)

Channel state information (CSI)-based human activity recognition (HAR) is vulnerable to performance degradation under domain shifts across varying physical environments. Continual learning (CL) offers a principled way to learn new domains sequentially while preserving past knowledge, but existing CL solutions for CSI-based HAR scale poorly with accumulating domains, rely on a large replay buffer, or incur linearly growing inference cost. In this letter, we propose Scene-Adaptive Mixture of Experts with Clustered Specialists (SAMoE-C), which formulates cross-domain CSI-based HAR as a mixture-of-experts system that enables scene-specific adaptation, via an attention-based semantic router that activates only selected experts for each input. Moreover, we develop a novel training protocol, which requires only a tiny replay buffer for stabilizing domain discrimination of the router. Experimental results on a four-scene CSI dataset demonstrate that SAMoE-C approaches the state-of-the-art accuracy, while maintaining a significantly lower inference cost. By jointly combining modular experts, selective activation with router and a lightweight training protocol, SAMoE-C enables scalable cross-domain CSI-based HAR deployment with low training overhead and high computational efficiency in real-world settings.