# H-Infinity Disturbance-Attenuating Latent Observer

- ID: 2939
- Canonical URL: https://synthcore.org/idea/2939/h-infinity-disturbance-attenuating-latent-observer
- API JSON: https://synthcore.org/api/idea/2939.json
- API Markdown: https://synthcore.org/api/idea/2939.md
- Verification status: failed_benchmark
- Source: [arXiv:2608.28349](https://arxiv.org/abs/2608.28349)
- Category: dynamics
- Solves: stability, accuracy
- ML areas: rnn, ssm, world-model, training-dynamics
- Math tags: control-theory, dynamical-systems, linear-algebra, optimization, statistics
- Ratings: usefulness 7/10; difficulty 7/10; novelty 7/10

## Idea description

Turn a latent recurrent model into an observer that continuously corrects its hidden state from noisy or partial observations while certifying both estimation-error convergence and disturbance attenuation. The bounded-real operator inequality becomes a trainable regularizer for a neural correction gain, providing a principled alternative to unconstrained teacher forcing or ad hoc residual correction.

## Mathematical statement

Let the latent plant be $\dot z=A_\phi(\theta,t)z+B_w w$ with observations $y=C_\phi z+D_w w$. Use the observer $\dot{\hat z}=f_\phi(\hat z,\theta,t)+K_\eta(\theta,t)(y-C_\phi\hat z)$ and error $e=z-\hat z$. Its local error dynamics are $\dot e=A_c e+B_cw$, where $A_c=A_\phi-K_\eta C_\phi$ and $B_c=B_w-K_\eta D_w$. For performance output $q=C_e e+D_e w$, let $P(\theta,t)=P^\top\succ0$ be a storage metric. The bounded-real differential LMI is $\begin{bmatrix}\dot P+A_c^\top P+PA_c+C_e^\top C_e+2\alpha P & PB_c+C_e^\top D_e\\ B_c^\top P+D_e^\top C_e & D_e^\top D_e-\gamma^2I\end{bmatrix}\preceq0$. It implies $\dot V+2\alpha V+\|q\|^2-\gamma^2\|w\|^2\le0$ for $V=e^\top Pe$, giving decay rate $\alpha$ and induced disturbance gain at most $\gamma$.

## Key formulas

- $$\dot{\hat z}=f_\phi(\hat z,\theta,t)+K_\eta(\theta,t)(y-C_\phi\hat z),\qquad e=z-\hat z.$$
- $$\dot e=A_c e+B_cw,\qquad A_c=A_\phi-K_\eta C_\phi,\qquad B_c=B_w-K_\eta D_w.$$
- $$\begin{bmatrix}\dot P+A_c^\top P+PA_c+C_e^\top C_e+2\alpha P & PB_c+C_e^\top D_e\\ B_c^\top P+D_e^\top C_e & D_e^\top D_e-\gamma^2I\end{bmatrix}\preceq0,\qquad P\succeq p_{\min}I.$$
- $$\dot V+2\alpha V+\|q\|^2-\gamma^2\|w\|^2\le0,\qquad V(e,t)=e^\top P(\theta,t)e.$$

## Implementation notes

1. Integration point: attach an observer head to a neural state-space model, world model, or recurrent encoder. The predictor evolves $\hat z$ autonomously, while each observation produces innovation $r_k=y_k-C_\phi\hat z_k$ and correction $K_\eta r_k$. Use diagonal or low-rank $P$ for large latent dimensions. 2. Pseudocode: predict $\hat z_{k+1}=\hat z_k+h f_\phi(\hat z_k,\theta_k)$; calculate $r_k$ and correct with $hK_\eta r_k$; estimate $A_c$ using Jacobian-vector products; construct $B_c,C_e,D_e$ from the assumed noise and output maps; evaluate the largest eigenvalue of the symmetric bounded-real LMI matrix and add its positive part to the loss. Parameterize $P=LL^\top+p_{\min}I$ and optimize $K_\eta$ jointly with the predictor. 3. Computed from the mechanism: the storage inequality and target gain $\gamma$; estimated empirically: local Jacobians, observation-noise statistics, and minibatch averages over parameter/time points. 4. First experiment: use a partially observed damped oscillator and Lorenz-63, followed by a small video latent model. Compare an equal-parameter GRU with teacher forcing, a learned Kalman-style correction, and this bounded-real observer; ablate the LMI while retaining the correction gain. Evaluate open-loop rollouts under observation noise and impulsive disturbances. 5. Pre-registered signature: with zero disturbance, $V(e_k)$ should decay at least as $e^{-2\alpha kh}V(e_0)$ after a discretization transient; with disturbances, cumulative output energy should satisfy $\sum_k\|q_k\|^2\le\gamma^2\sum_k\|w_k\|^2+V_0$. Reject the transfer if measured induced gain exceeds $1.2\gamma$, or if the estimated LMI is feasible but held-out error decay fails.

## Verification

- Status: failed_benchmark
- Mechanism evidence: yes
- Mechanism confirmed: no
- Practical verdict: no_effect
- Verdict: Built a scalar H-infinity latent observer MVP with numerical storage-metric search, bounded-real LMI evaluation, exact disturbance-transfer verification, and time-domain dissipation checks. The feasible observer gain K=1.7804 had LMI maximum eigenvalue -0.00098, exact H-infinity gain 0.8166 below gamma=0.9, and zero-disturbance V decay rate 4.36 versus the required 0.4. The unconstrained high-gain baseline K=10 converged faster but was not LMI-feasible and exceeded the target gain with exact H-infinity gain 0.9615, demonstrating the claimed stability/attenuation tradeoff on this toy system.

### Mechanism check

- Verdict: Built a scalar H-infinity latent observer MVP with numerical storage-metric search, bounded-real LMI evaluation, exact disturbance-transfer verification, and time-domain dissipation checks. The feasible observer gain K=1.7804 had LMI maximum eigenvalue -0.00098, exact H-infinity gain 0.8166 below gamma=0.9, and zero-disturbance V decay rate 4.36 versus the required 0.4. The unconstrained high-gain baseline K=10 converged faster but was not LMI-feasible and exceeded the target gain with exact H-infinity gain 0.9615, demonstrating the claimed stability/attenuation tradeoff on this toy system.
- Confidence: 8/10
- Limitations: Only a scalar linear continuous-time error system was tested; no neural predictor, learned gain, GRU comparison, Lorenz-63 system, partial-observation training, parameter/time-varying P, or large-scale computational overhead was evaluated. The gain and storage metric were selected by grid search rather than jointly learned.

### Practical benchmark

- Paired seeds: 8
- Baseline mean: 0
- Idea mean: 0
- p-value: 1
- Paired wins: 0/8
- Benchmark verdict: no measurable effect

## Artifacts

- [bench_report.json](https://synthcore.org/code/1139/bench_report.json)
- [observer_experiment.py](https://synthcore.org/code/1139/observer_experiment.py)
- [report.md](https://synthcore.org/code/1139/report.md)
- [report_bench_2026-09-02T122319.md](https://synthcore.org/code/1139/report_bench_2026-09-02T122319.md)
- [results.json](https://synthcore.org/code/1139/results.json)
- [stage2_bench.py](https://synthcore.org/code/1139/stage2_bench.py)
- [Download all files as ZIP](https://synthcore.org/download/1139)

## Disclaimer

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