Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.00458 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912866029797376 |
|---|---|
| author | Haq, Omer |
| author_facet | Haq, Omer |
| contents | We introduce LatentTrack (LT), a sequential neural architecture for online probabilistic prediction under nonstationary dynamics. LT performs causal Bayesian filtering in a low-dimensional latent space and uses a lightweight hypernetwork to generate predictive model parameters at each time step, enabling constant-time online adaptation without per-step gradient updates.
At each time step, a learned latent model predicts the next latent distribution, which is updated via amortized inference using new observations, yielding a predict--generate--update filtering framework in function space. The formulation supports both structured (Markovian) and unstructured latent dynamics within a unified objective, while Monte Carlo inference over latent trajectories produces calibrated predictive mixtures with fixed per-step cost. Evaluated on long-horizon online regression using the Jena Climate benchmark, LT consistently achieves lower negative log-likelihood and mean squared error than stateful sequential and static uncertainty-aware baselines, with competitive calibration, demonstrating that latent-conditioned function evolution is an effective alternative to traditional latent-state modeling under distribution shift. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_00458 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LatentTrack: Sequential Weight Generation via Latent Filtering Haq, Omer Machine Learning Artificial Intelligence Robotics We introduce LatentTrack (LT), a sequential neural architecture for online probabilistic prediction under nonstationary dynamics. LT performs causal Bayesian filtering in a low-dimensional latent space and uses a lightweight hypernetwork to generate predictive model parameters at each time step, enabling constant-time online adaptation without per-step gradient updates. At each time step, a learned latent model predicts the next latent distribution, which is updated via amortized inference using new observations, yielding a predict--generate--update filtering framework in function space. The formulation supports both structured (Markovian) and unstructured latent dynamics within a unified objective, while Monte Carlo inference over latent trajectories produces calibrated predictive mixtures with fixed per-step cost. Evaluated on long-horizon online regression using the Jena Climate benchmark, LT consistently achieves lower negative log-likelihood and mean squared error than stateful sequential and static uncertainty-aware baselines, with competitive calibration, demonstrating that latent-conditioned function evolution is an effective alternative to traditional latent-state modeling under distribution shift. |
| title | LatentTrack: Sequential Weight Generation via Latent Filtering |
| topic | Machine Learning Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2602.00458 |