Integrating Feature Selection in Counterfactual Generation: An Optimized Framework Using Simulated Annealing
Bibliographic record
Abstract
In this study, we introduce an innovative framework for counterfactual generation that strategically modifies key “actionable features” to alter decision outcomes. Utilizing Simulated Annealing (SA), our approach optimizes the counterfactual generation process by aiming to reduce both the number and magnitude of feature modifications required, thereby ensuring the generated counterfactuals are sparse, proximate, and valid. To prioritize feature modifications, we apply Local Interpretable Model-agnostic Explanations (LIME) and its variant, S-LIME. These model-agnostic methods help interpret the importance of features by building understandable models for individual data instances, clarifying each feature's influence on the model's decision. Additionally, we develop a novel adaptive Gaussian perturbation technique to produce candidate counterfactuals close to the original input, ensuring their relevance by progressively perturbing the vicinity of the input. A notable advancement of our method over existing techniques is its ability to address multiclass predictions, expanding its applicability beyond the binary prediction limitations of prior works using LIME. Our comprehensive evaluation of the proposed method across diverse datasets-ranging from population growth and malware detection to house pricing-includes detailed experiments and comparisons with leading approaches. The promising results underscore the potential of our approach to significantly enhance model interpretability and decision-making transparency in various applications.
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 | 0.010 | 0.032 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 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".