Application of Scientific Machine Learning Methods in Epidemical Modeling: Undetected COVID-19 Population Ratio Prediction
Bibliographic record
Abstract
Combining deep learning techniques with mathematical models can compensate for the drawbacks of each method. Deep learning methods can help increase accuracy, while mathematical models can save computational costs by constraining the neural network (NN) structure. The combined method is efficient for capturing the dynamics of real-life data with a small sample size. In addition, the trained model can be sparsely regressed to a concise parametric model using data-driven methods for learning the mechanics, and used for prediction. In this thesis we focus on the application to epidemic modelling. Some theoretical background is introduced in Chapter 1. In Chapter 2, the methodology for developing scientific learning model on epidemic population data is explained. Discrete time-shifting is considered because of the latent detection of infected population. The undetected rate is modelled in two ways: using a constant detection rate, and by involving the detection rate as an output of the NN. The trained models are regressed to the dynamic systems via SINDy for learning the terms, and for prediction. In Chapter 3, we apply the model to the first wave of COVID-19 in Calgary where the data only contains 12 points of weekly detected infected population. In Chapter 4, the model is extended to the first wave of COVID-19 in Canada, where there is training data containing 18 weeks of the detected infected population and the deceased population. In Chapter 5, some conclusions are drawn and the future potential of this method is discussed.
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 | 0.003 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| 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) | 0.001 | 0.000 |
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".