Online Learning via Learned Latent Bayesian Tracking
cs.LG, eess.SP
Submitted: 2026-09-25
Updated: 2026-09-28
Comments: Accepted at NeurIPS 2026
License: http://creativecommons.org/licenses/by/4.0/
The gist: Online learning in non-stationary environments requires models to adapt rapidly from streaming data under strict computational constraints.
Terminology
Abstract
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.
Sources
- Low-rank extended Kalman filtering for online learning of neural networks from streaming data
- KOALA++: Efficient Kalman-Based Optimization with Gradient-Covariance Products
- An Overview of the 3GPP Study on Artificial Intelligence for 5G New Radio
- Evaluating Prediction-Time Batch Normalization for Robustness under Covariate Shift
- MNIST-C: A Robustness Benchmark for Computer Vision
- Wide Residual Networks
- Variational Continual Learning
- AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
- RobustBench: a standardized adversarial robustness benchmark
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks