A Novel Ship Design Optimization Framework Using Fine-Tuned and Reward-Directed Diffusion Model
Bibliographic record
Abstract
Abstract Optimizing ship hull design using existing datasets poses significant challenges in the marine industry. This study presents an innovative framework for generative artificial intelligence (AI) that uses fine-tuned and reward-weighted sampling during inference in a diffusion model to generate ship hull designs with reduced resistance. We use a parametric approach for the design representation of the ship hull and apply the developed framework to generate new parameters that represent a hull with reduced resistance. Empirical results indicate that the reward guidance substantially improves the diffusion model’s ability to produce samples with reduced resistance in ship design generation tasks. A specific advantage of the reward-directed approach is its effectiveness when engineering simulations are needed to compute performance metrics, such as resistance. This makes the objectives non-differentiable and thus challenging for traditional gradient-based optimization techniques. We first show that the diffusion model can generate 3D ship designs within complex simulation environments. Then, we demonstrate that our framework successfully generates high-performance ship designs that meet engineering criteria directly from tabular data. This work introduces the use of reward guidance for a Markov decision process (MDP), providing an intuitive approach to provide directional sampling in diffusion models, particularly in complex and nondifferentiable settings. This study is a steppingstone towards constrained-based design optimization using generative AI for engineering applications. Using this approach, performance and operational considerations will be embedded in the optimization process during sampling and reward augmentation.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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 teacher head, 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".