Short-term and Long-term Causal Effect Estimation with Double-head Transformer
Bibliographic record
Abstract
The study of long-term effects, especially when they differ significantly from short-term implications, is paramount in understanding the broader implications of any condition. Estimating these long-term causal effects is not just crucial but is also a pre-requisite for a comprehensive evaluation and effective decision-making in various domains, from healthcare to economics. Most contemporary models, equipped with advanced algorithms, have been able to successfully estimate these long-term effects using short-term surrogates, primarily based on the surrogate assumption. However, it's evident that challenges persist, especially concerning the long-term dependency and the causal structure of the data. Addressing these issues is imperative for enhancing the accuracy and reliability of the estimations. In light of these challenges, we introduce TransLTEE, a novel double-heads Transformer model. This model aims to rectify the vanishing gradient problem frequently encountered in the long-term treatment effect estimation with double-head RNNs. By strategically replacing RNNs with Transformers, TransLTEE offers a more robust solution. The inclusion of encoder and decoder blocks in TransLTEE is a deliberate design choice that facilitates learning the distribution distance between treatment groups and predicted outcomes at each timestep. This ensures a more precise estimation of both short-term and long-term causal effects, potentially revolutionizing the way we understand and interpret data.
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 | 0.006 | 0.018 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.002 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.007 | 0.001 |
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".