# Model-Ensemble Space-Filling Explorer - ID: 2237 - Canonical URL: https://synthcore.org/idea/2237/model-ensemble-space-filling-explorer - API JSON: https://synthcore.org/api/idea/2237.json - API Markdown: https://synthcore.org/api/idea/2237.md - Verification status: mechanism_failed - Source: [arXiv:2608.13360](https://arxiv.org/abs/2608.13360) - Category: training - Solves: sample-efficiency, accuracy, stability - ML areas: world-model, rl, optimizer, data-augmentation - Math tags: control-theory, optimization, probability, dynamical-systems, statistics - Ratings: usefulness 7/10; difficulty 6/10; novelty 7/10 ## Idea description Train an input-generation policy or differentiable signal parameterization to produce trajectories that cover the joint input-state feature space while remaining informative for every plausible neural world model. Replace single-model experiment design by an expectation over an ensemble of models, and optimize this objective with stochastic model and trajectory samples. ## Mathematical statement The paper models a trajectory through x_(k+1) = f_m(x_k,u_k) and y_k = g_m(x_k,u_k), with feature z_k = (u_k,x_k). Let M be an ensemble or posterior distribution of candidate world models, let psi parameterize an input policy u_k = pi_psi(x_k,xi_k), and let z_(0:T)^m(psi) be the resulting trajectory under model m sampled from p(m). Define a space-filling or information criterion Phi(psi,m) = log det(epsilon I + (1/(T+1)) sum_k w(z_k) w(z_k)^T), where w(z) is a bounded feature embedding and epsilon is a positive numerical stabilizer. The robust design objective is J(psi) = E_m,xi[Phi(psi,m)] - lambda C(psi), where C penalizes input magnitude, constraint violations, or unsafe states. Stochastic approximation uses psi_(r+1) = psi_r + eta_r times an estimated gradient of J. If the criterion and gradient variance are bounded, diminishing step sizes with sum eta_r infinite and sum eta_r squared finite provide the standard stochastic-approximation convergence regime toward a stationary point. ## Key formulas - x_(k+1) = f_m(x_k,u_k), y_k = g_m(x_k,u_k), z_k = (u_k,x_k). - J(psi) = E_(m sampled from p(m), xi)[Phi(z_(0:T)^m(psi))] - lambda C(psi). - Phi = log det(epsilon I + (1/(T+1)) sum_(k=0)^T w(z_k) w(z_k)^T). - psi_(r+1) = psi_r + eta_r [ (1/B) sum_(b=1)^B grad_psi Phi(z_(0:T)^(m_b)(psi_r)) - lambda grad_psi C(psi_r) ], with m_b sampled independently from p(m). ## Implementation notes 1. Integration point: use this as a data-collection module before or between updates of a neural state-space model. Parameterize the experiment by a small MLP policy u_k = pi_psi(x_k,xi_k), or by bounded Fourier or chirp coefficients if differentiating through the environment is unavailable. Maintain an ensemble of M world models f_(theta_m) trained on the currently collected data. 2. Pseudocode: sample B model indices; roll out the policy for T steps with x_(k+1) = f_(theta_m)(x_k,u_k); compute z_k = (u_k,x_k); construct G_m = epsilon I + (1/(T+1)) sum_k w(z_k)w(z_k)^T; set the minibatch objective to the average of log det(G_m) minus lambda times the safety and action cost; backpropagate through the rollout and update psi by gradient ascent. Clip actions and assign a large penalty to unsafe states. 3. Computed from the paper's mechanism: sampling a population of candidate models, optimizing the expected criterion, and using stochastic approximation. Estimated empirically: the model distribution, gradients, feature map w, and safety penalty. 4. First cheap experiment: use the scalar nonlinear system x_(k+1) = 0.8 x_k + a x_k cubed + u_k, with an ensemble in which coefficient a varies, and compare random excitation, single-model design, and ensemble design while training small neural predictors. Measure held-out one-step and 20-step prediction error, minimum-model log determinant, and gradient variance. The quantitative prediction is that ensemble design improves the minimum-model log determinant by at least 10 percent over single-model design at equal rollout budget, while the variance of the minibatch objective estimate decreases approximately as 1/B. As ensemble spread increases, single-model performance should degrade sharply, whereas robust design should retain flatter worst-model information until input constraints become active. ## Verification - Status: mechanism_failed - Mechanism evidence: yes - Mechanism confirmed: no - Verdict: Built a differentiable ensemble space-filling explorer with nonlinear candidate world models, log-determinant feature coverage, nominal and ensemble optimizers, finite-difference validation, ensemble-spread sweeps, and minibatch-gradient variance tests. The gradient check passed to 2.36e-9 maximum absolute error, and gradient MSE followed approximately 1/B with B*MSE near 4.8e-8. However, robust optimization did not improve worst-model log determinant: at spread 0.18 the change was -0.026%, far below the proposed 10% gain, so the central effect was not observed. ### Mechanism check - Verdict: Built a differentiable ensemble space-filling explorer with nonlinear candidate world models, log-determinant feature coverage, nominal and ensemble optimizers, finite-difference validation, ensemble-spread sweeps, and minibatch-gradient variance tests. The gradient check passed to 2.36e-9 maximum absolute error, and gradient MSE followed approximately 1/B with B*MSE near 4.8e-8. However, robust optimization did not improve worst-model log determinant: at spread 0.18 the change was -0.026%, far below the proposed 10% gain, so the central effect was not observed. - Confidence: 9/10 - Limitations: Only a small open-loop scalar toy system and one fixed initialization were tested; no neural predictor training, learned policy, safety constraints, held-out prediction error, multiple random seeds, GPU execution, or larger model ensembles were evaluated. ## Artifacts - [ensemble_explorer.py](https://synthcore.org/code/888/ensemble_explorer.py) - [report.md](https://synthcore.org/code/888/report.md) - [results.json](https://synthcore.org/code/888/results.json) - [Download all files as ZIP](https://synthcore.org/download/888) ## Disclaimer AI-generated research hypothesis, automatically tested. Not peer-reviewed.