# Small-gain-certified surrogate controller

- ID: 266
- Canonical URL: https://synthcore.org/idea/266/small-gain-certified-surrogate-controller
- API JSON: https://synthcore.org/api/idea/266.json
- API Markdown: https://synthcore.org/api/idea/266.md
- Verification status: beats_baseline
- Source: [arXiv:2607.04362](https://arxiv.org/abs/2607.04362)
- Category: training
- Solves: stability, accuracy, generalization
- ML areas: world-model, ssm, rl
- Math tags: control-theory, dynamical-systems, functional-analysis, approximation-theory
- Ratings: usefulness 7/10; difficulty 5/10; novelty 7/10

## Idea description

Train a neural controller as a uniformly accurate surrogate of a trusted but expensive controller, and use a measured small-gain condition to decide whether the surrogate is safe for closed-loop deployment. The approximation tolerance becomes an interpretable residual-state budget instead of an opaque validation metric.

## Mathematical statement

Theorem 2 guarantees a neural-operator surrogate \(\widehat{\mathcal U}_N\) with uniform error \(|\mathcal U_N(\mathbf f_N,u)-\widehat{\mathcal U}_N(\mathbf f_N,u)|\leq\varepsilon\) over the compact plant/state input class. Theorem 3 assumes a local feedback Lipschitz constant \(\mathrm{Lip}_K(r_0)<1\), where \(r_0\) is the allowed state radius, and requires \(r_0<\bar r=\min(r^*,\rho_f)\), with \(r^*=\mathrm{Lip}_K^{-1}(1)\) and \(\rho_f\) the plant's admissible basin radius. Under these assumptions, the closed loop is practically stable and its residual class-\(\mathcal K\) bound \(\gamma(\varepsilon)\) scales linearly with the uniform surrogate error \(\varepsilon\). Here \(K\) denotes the closed-loop feedback map, \(\mathrm{Lip}_K(r_0)\) is its local state Lipschitz constant, and \(\gamma\) maps controller approximation error to the final state residual. The engineering principle is to minimize both approximation error and local feedback gain, then validate the inequality before deployment.

## Key formulas

- $$\left|\mathcal{U}_{N}(\mathbf{f}_{N},u)-\widehat{\mathcal{U}}_{N}(\mathbf{f}_{N},u)\right|\leq\varepsilon,\qquad \forall(\mathbf f_N,u)\in\mathcal F_N^{\star}(B_f,L_f)\times\mathcal V(R,L_u),$$
- $$\mathrm{Lip}_{K}(r_0)<1,\qquad r_0<\bar r:=\min(r^*,\rho_f),\qquad r^*:=\mathrm{Lip}_{K}^{-1}(1),$$
- $$\|u(t)\|_{\infty}\ \text{eventually enters a residual ball of radius}\ \gamma(\varepsilon),\qquad \gamma(\varepsilon)=O(\varepsilon).$$
- $$\widehat{\mathcal U}_{\mathrm{deploy}}(f,u)=\begin{cases}\widehat{\mathcal U}_N(f,u),&\widehat{\mathrm{Lip}}_K(r_0)<1\ \text{and}\ \|u\|_\infty\le r_0,\\\mathcal U_N(f,u)\ \text{or a conservative fallback},&\text{otherwise.}\end{cases}$$

## Implementation notes

Use a high-fidelity controller, optimizer, or solver-derived policy as the teacher and a neural operator as the deployable student. The integration point is the action computation from a plant descriptor \(f\) and sampled state \(u\); no kernel PDE or teacher solve should be needed during normal inference. Generate training examples by sampling plant coefficients and closed-loop states inside a declared radius \(r_0\), evaluate the teacher action, and minimize supervised imitation error. At each validation checkpoint, estimate \(\hat\varepsilon\) as the maximum absolute action error over dense random samples, adversarially perturbed inputs, and rollout-generated states. Estimate the local feedback gain with finite differences: for states \(u_i\) satisfying \(\|u_i\|_\infty\le r_0\), draw normalized perturbations \(v_i\), compute \(g_i=|\hat U(f_i,u_i+\delta v_i)-\hat U(f_i,u_i)|/(\delta\|v_i\|_\infty)\), and set \(\widehat{\mathrm{Lip}}_K(r_0)=\max_i g_i\). Add \(\alpha\max(0,g_i-\ell)^2\) to the loss for a target gain \(\ell<1\). The paper supplies the small-gain criterion and predicts a residual radius linear in \(\varepsilon\); the finite-sample error and gain are empirical estimates. Start with a 1D hyperbolic PDE discretized into 32--128 cells and compare the teacher, an unconstrained surrogate, and the gain-regularized surrogate. Plot final state residual versus measured \(\hat\varepsilon\), and measure inference latency and unstable-rollout rate. Success means substantially lower inference cost, no unstable rollouts inside the validated region, and an approximately linear residual-error relationship.

## Verification

- Status: beats_baseline
- Mechanism evidence: yes
- Mechanism confirmed: no
- Practical verdict: helps
- Verdict: Built a scalar small-gain-certified surrogate-controller MVP with teacher imitation, finite-difference gain estimation, gain regularization, rollouts, and reproducible sweeps. The mechanism manifested: strict |q|>1 cases grew, |q|<1 cases contracted, the marginal |q|=1 case was not asymptotically stable, and residuals matched epsilon/(1-|q|) to 1.9e-14 relative error across gains. The neural surrogate achieved estimated gain below 1 and zero unstable rollouts, but gain regularization did not improve the rollout metric in this run.

### Mechanism check

- Verdict: Built a scalar small-gain-certified surrogate-controller MVP with teacher imitation, finite-difference gain estimation, gain regularization, rollouts, and reproducible sweeps. The mechanism manifested: strict |q|>1 cases grew, |q|<1 cases contracted, the marginal |q|=1 case was not asymptotically stable, and residuals matched epsilon/(1-|q|) to 1.9e-14 relative error across gains. The neural surrogate achieved estimated gain below 1 and zero unstable rollouts, but gain regularization did not improve the rollout metric in this run.
- Confidence: 8/10
- Limitations: This is a scalar discrete-time toy analogue, not the paper's 1D hyperbolic PDE or neural operator. The empirical gain is finite-sample and finite-difference estimated rather than a formal uniform certificate; no adversarial validation, teacher-solver cost comparison, fallback controller, or statistically repeated seeds were tested.

### Practical benchmark

- Paired seeds: 8
- Baseline mean: 0
- Idea mean: 0
- p-value: 0.0081
- Paired wins: 8/8
- Benchmark verdict: idea better (significant)

## Artifacts

- [bench_report.json](https://synthcore.org/code/109/bench_report.json)
- [report.md](https://synthcore.org/code/109/report.md)
- [report_bench_2026-08-30T142503.md](https://synthcore.org/code/109/report_bench_2026-08-30T142503.md)
- [results.json](https://synthcore.org/code/109/results.json)
- [small_gain_experiment.py](https://synthcore.org/code/109/small_gain_experiment.py)
- [stage2_small_gain_bench.py](https://synthcore.org/code/109/stage2_small_gain_bench.py)
- [Download all files as ZIP](https://synthcore.org/download/109)

## Disclaimer

AI-generated research hypothesis, automatically tested. Not peer-reviewed.