PEA: Parallel Evolutionary Algorithm by Separating Convergence and Diversity for Large-Scale Multi-Objective Optimization
Bibliographic record
Abstract
Running evolutionary algorithms in parallel is an intuitive way to speed up the process of solving large-scale multi-objective optimization problems, which have hundreds or thousands of decision variables. However, the framework of the existing multi-objective evolutionary algorithms seriously limits their parallelization. During each iteration, the environmental selection operators present in the existing framework need to collect and compare all the candidate solutions to balance the convergence and diversity, thus dividing the whole evolutionary process into a series of dependent sub-processes and resulting in frequent data transmission. To address this issue, we propose a novel parallel framework that separates the environmental selection operator from the entire evolutionary process, evidently removing the dependencies among sub-processes and reducing the data transmission. On the basis of the parallel framework, a new parallel evolutionary algorithm, namely PEA, is designed. In PEA, the convergence is achieved by a series of independent sub-populations, and the diversity is merely emphasized at the converged solutions from each subpopulation, which is helpful for avoiding that the environmental selection operator limits the parallelization of the algorithm. Moreover, a new environmental selection strategy is proposed to improve the diversity without considering the convergence. To assess the performance of the proposed PEA, we compare it with five representative multi-objective evolutionary algorithms in terms of both the convergence and diversity. The performance of the parallel framework is also analyzed by comparing with two existing parallel models. The experimental results demonstrate the superiority of the proposed parallel algorithms in terms of the convergence, diversity, and speedup.
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 | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".