Metraipolis › archive › dc_53e5f2ea0ca8

Online Learning via Learned Latent Bayesian Tracking — arXiv:2609.31559

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.31559
License (as archived)
CC0-1.0
Archived
2026-10-08T02:13:20.349Z
SHA-256 of the archived text
2ac8648346bc27e1d6afd2843e02efe81d3dfb3915932152ff7037a63724f291
Stable URL of this page
https://metraipolis.com/documento/dc_53e5f2ea0ca8

How to cite

Online Learning via Learned Latent Bayesian Tracking — arXiv:2609.31559 — arXiv (en), https://arxiv.org/abs/2609.31559 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_53e5f2ea0ca8 (Metraipolis archive dc_53e5f2ea0ca8, sha256 2ac8648346bc27e1d6afd2843e02efe81d3dfb3915932152ff7037a63724f291)
BibTeX
@misc{dc53e5f2ea0ca8,
  title = {Online Learning via Learned Latent Bayesian Tracking — arXiv:2609.31559},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.31559},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_53e5f2ea0ca8; sha256 2ac8648346bc27e1d6afd2843e02efe81d3dfb3915932152ff7037a63724f291},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31559. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Online learning in non-stationary environments requires models to adapt rapidly from streaming data under strict computational constraints. A principled approach casts online learning as Bayesian state tracking, where model parameters are updated sequentially via Bayesian filtering. However, applying Bayesian filters directly to modern deep models is computationally prohibitive due to the high dimensionality of parameter space, forcing existing methods to rely on restrictive approximations or manually designed low-dimensional subspaces. In this work, we identify the absence of a suitable low-dimensional dynamical representation as the core bottleneck in Bayesian filtering-based online learning. Accordingly, we propose Adaptive Update through Representation Adaptation (AURA), a meta-learning framework that learns offline a low-dimensional latent state-space model governing the evolution of optimal model parameters under distribution shift. Online adaptation is then performed via extended Kalman filtering in this learned latent space followed by reconstruction of the full model parameters through a learned lifting map, enabling efficient single-step online adaptation while preserving model expressiveness. Evaluated on online adaptation of neural wireless receivers under time-varying channels and on non-stationary image classification, AURA shows substantial improvements in adaptation speed, accuracy, and computational efficiency over existing online learning and Bayesian filtering baselines, demonstrating that an adaptation-aware latent geometry is beneficial for effective Bayesian online learning in high-dimensional models.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Signal Processing (eess.SP).
Pagina del lavoro: https://arxiv.org/abs/2609.31559 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T01:43:06.758Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).