{
 "artifacts": null,
 "category": "dynamics",
 "description": "Insert a constraint-aware observer between a neural state-space transition and its next prediction. The observer propagates latent event times, incorporates partial observations, and projects the result onto the set satisfying both lower-bound causality and upper-bound token-lifetime constraints, preventing impossible latent trajectories from entering the recurrent model.",
 "formulas_latex": [
  "$$x=x_{0}\\oplus Ax\\oplus Bu\\oplus Rw,\\qquad y=Cx,\\qquad x\\geq Ux.$$",
  "$$\\tilde z_{k+1}=F_{\\theta}(z_k,u_k),\\qquad \\hat z_{k+1}=\\max\\left\\{z:\\ z\\leq\\tilde z_{k+1},\\ z\\geq Uz,\\ Hz=y_{k+1}\\right\\},$$",
  "$$\\hat z_{k+1,i}\\leq\\tilde z_{k+1,i},\\qquad \\hat z_{k+1,i}\\geq\\max_j\\left(U_{ij}+\\hat z_{k+1,j}\\right),$$",
  "$$\\Delta_i=\\max_j(U_{ij}+\\tilde z_{k+1,j})-\\tilde z_{k+1,i};\\qquad \\Delta_i\u003e0\\Longrightarrow\\text{the unconstrained prediction violates an upper-bound window by at least }\\Delta_i.$$"
 ],
 "id": 1668,
 "implementation": "1. Integration point: use a recurrent state-space model or world model whose latent vector contains timestamp-like coordinates, such as object arrival times, contact times, queue events, or discretized latent milestones. Keep the learned transition $F_\\theta$ unchanged initially. After every transition prediction, apply an observer/projection layer before feeding the state to the next step. Known observation coordinates are clamped to measured event times; unknown coordinates are corrected. 2. Pseudocode: `z_pred = F_theta(z, u)`; compute the constraint residual `r[i] = max_j(U[i,j] + z_pred[j]) - z_pred[i]`; for violated constraints, solve the small linear program `maximize sum_i z[i]` subject to `z[i] \u003c= z_pred[i]`, `z[i] \u003e= U[i,j] + z[j]`, and `z[observed] = y`. For a cheaper approximation, use sparse relaxation sweeps over the difference-constraint graph, lowering violated coordinates and restoring observed coordinates after each sweep. 3. Computed from the paper: monotone propagation, the inequality $z\\geq Uz$, and greatest-subapproximation correction. Estimated empirically: the neural transition, observation noise, constraint matrix $U$, and projection convergence. 4. First experiment: train a GRU state-space model on synthetic partially observed timed event graphs with random lower bounds and windows $[l_{ij},u_{ij}]$. Compare vanilla rollout, observation-only correction, and the projection observer. Measure impossible-window violations, hidden-event MAE, and error after each observation. The prediction is that projected rollouts have zero constraint violations up to solver tolerance, while vanilla violations grow with horizon. After an observation touching a connected component, hidden-state error should decrease over subsequent propagation steps. Sweep window tightness and identify the infeasibility threshold; before that threshold, test for at least a 20% reduction in long-horizon MAE.",
 "math_summary": "The paper models dater vectors with $x=x_{0}\\oplus Ax\\oplus Bu\\oplus Rw$, observations $y=Cx$, and P-Time Event Graph constraints $x\\geq Ux$. Here $\\oplus$ is maximum, multiplication is addition, $\\varepsilon=-\\infty$, and $e=0$; $x_i(k)$ is the firing time of latent transition $i$ at event index $k$. Matrix $A$ propagates causal lower-bound constraints, while $U$ encodes upper-bound token-lifetime restrictions. For a neural model, let $z_k$ be latent event times, $F_\\theta$ the learned transition, and $H$ the observation map. First calculate $\\tilde z_{k+1}=F_\\theta(z_k,u_k)$, then choose the greatest feasible under-approximation $\\hat z_{k+1}\\leq\\tilde z_{k+1}$ satisfying $\\hat z_{k+1}\\geq U\\hat z_{k+1}$ and $H\\hat z_{k+1}=y_{k+1}$. Monotonicity is the useful mechanism: max-plus propagation preserves order, while projection removes estimates that violate temporal windows.",
 "math_tags": [
  "control-theory",
  "dynamical-systems",
  "optimization",
  "linear-algebra"
 ],
 "ml_areas": [
  "ssm",
  "rnn",
  "world-model",
  "training-dynamics"
 ],
 "paper": {
  "arxiv_id": "2608.01371",
  "arxiv_url": "https://arxiv.org/abs/2608.01371",
  "summary_what_math_gives_to_ml": "The paper provides a constraint-aware Luenberger observer for P-Time Event Graphs, where latent firing times are reconstructed from partial observations while enforcing both causal lower bounds and token-lifetime upper bounds. Its transferable asset is the construction of a greatest feasible sub-approximation: estimates are propagated through a recurrent max-plus model and corrected whenever newly observed outputs make earlier latent states inconsistent. This can be transferred to neural state-space or world-model inference as a hard or differentiable projection layer that prevents latent trajectories from violating known temporal-window constraints. The falsifiable prediction is that constraint projection eliminates physically impossible rollouts and that estimation error contracts after informative observations.",
  "title": "A Luenberger Observer for P-Time Event Graphs",
  "year": "2026"
 },
 "ratings": {
  "difficulty": 6,
  "novelty": 8,
  "usefulness": 6
 },
 "solves": [
  "stability",
  "accuracy"
 ],
 "title": "Temporal-Window Luenberger Projection",
 "url": "https://synthcore.org/idea/1668/temporal-window-luenberger-projection",
 "verification": {
  "peer_reviewed": false,
  "status": "unverified",
  "status_label": "Unverified",
  "verdict_source": "deterministic test code (paired-seed permutation statistics)",
  "verification_axes": {
   "benchmark_mechanism": {
    "confirmed": null,
    "tested": false
   },
   "practical_benchmark": {
    "beats_baseline": null,
    "tested": false
   },
   "toy_mechanism_gate": {
    "confirmed": null,
    "tested": false
   }
  }
 }
}