Teaching AI to Design From Humans: a Comparison of Behavioral Cloning Architectures
Bibliographic record
Abstract
Abstract Reinforcement Learning (RL) has created agents with superhuman performance in robotics and gaming. A central issue in using RL methods to automate engineering design is the inability to generalize and slow training. Even the most advanced curiosity-based RL algorithms require exploring millions of design states, which is infeasible with expensive physics models. Data from a human-subject design study shows that even novice human designers can solve design tasks in a few hundred actions. Behavioral cloning allows RL agents to imitate the policies of a human designer from their decision data. We evaluate the performance of a behavioral cloning agent trained on human design decision data collected in a controlled experiment. We compare three popular sequence learning architectures for behavioral cloning. Subsequently, we evaluate an AI design agent trained through behavioral cloning on human design decision data to automatically design an electric aircraft, starting from a baseline design. The results demonstrate that behavioral cloning effectively transfers human strategies to AI design agents with high sample efficiency.
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) | 0.000 | 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".