Bibliographic record
Abstract
This chapter covers the Bayesian filtering algorithms that deploy deep learning to build state-space models from raw data. Variational inference and amortized variational inference are reviewed, which are used to estimate an approximate posterior through reformulating the inference problem as an optimization problem aimed at minimizing the Kullback–Leibler divergence between the true and the approximate posteriors. A number of deep learning-based filtering algorithms are inspired by variational autoencoders. Such filters are trained by optimizing the evidence lower bound. The presented deep learning-based filtering algorithms include deep Kalman filter, backpropagation Kalman filter, differentiable particle filter, deep Rao–Blackwellized particle filter, deep variational Bayes filter, and Kalman variational autoencoder. Then, deep variational information bottleneck is reviewed, which aims at providing an optimal representation in terms of a trade-off between complexity of the representation and its predictive power. The issue of robustness is discussed by presenting the Wasserstein distributionally robust Kalman filter. Hierarchical invertible neural transport is presented, which can provide both the joint and the conditional densities. The reviewed applications of deep learning-based filters include predicting the effect of anti-diabetic drugs based on the electronic health records and autonomous driving using the KITTI Vision Benchmark.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.013 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".