△ Mechanism confirmed, baseline not beaten 2026

Event-triggered phase desynchronisation for recurrent hidden states

Implementation & benchmark of arXiv:2608.14907 — Event-Triggered Stabilisation of Desynchronisation in Networked Oscillatory Systems

Usefulness7/10

Difficulty6/10

Novelty8/10

Dynamics Stability Accuracy Scalability Rnn Ssm Training dynamics Regularization Dynamical systems Control theory Geometry Stochastic processes

Source paper: Event-Triggered Stabilisation of Desynchronisation in Networked Oscillatory Systems arXiv:2608.14907 ⓘ · analyzed Sep 1, 2026

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 Not peer-reviewed.

Idea description

Augment each recurrent or state-space hidden channel with a two-dimensional oscillatory state and periodically compute a pseudo-phase from its Cartesian coordinates. Use sparse event-triggered feedback to reduce the squared phase order parameter, preventing hidden channels from synchronising while avoiding the computation and communication cost of continuously recomputing the control signal. The controller acts as a tangent rotation of each two-dimensional hidden state, changing phase diversity without directly changing channel amplitude.

Formulas

$$\hat\theta_j=\operatorname{atan2}(y_j,x_j),\qquad z_j=e^{i\hat\theta_j},\qquad r=\frac{1}{N}\sum_{j=1}^{N}z_j,\qquad V(\hat\theta)=|r|^2.$$

$$\frac{\partial V}{\partial\hat\theta_j}=-\frac{2}{N}\operatorname{Im}(z_j\bar r),\qquad u_j=-k\frac{\partial V}{\partial\hat\theta_j}=\frac{2k}{N}\operatorname{Im}(z_j\bar r),\qquad \dot V=-k\|\nabla_{\hat\theta}V\|_2^2.$$

$$\dot h_j=F_j(h,x;\theta)+u_jJh_j,\qquad J=\begin{bmatrix}0&-1\\1&0\end{bmatrix},\qquad u(t)=u(t_\ell)\quad\text{for }t\in[t_\ell,t_{\ell+1}),$$

$$\|u(t)-u(t_\ell)\|_2\geq\delta\quad\text{triggers an event},\qquad t_{\ell+1}-t_\ell\geq\frac{\delta}{M}\quad\text{if }\|\dot u(t)\|_2\leq M.$$

Mathematical statement

For hidden oscillator state $h_j=(x_j,y_j)\in\mathbb{R}^2$, define the pseudo-phase $\hat\theta_j=\operatorname{atan2}(y_j,x_j)$ and complex phase $z_j=e^{i\hat\theta_j}$. The synchrony order parameter is $r=N^{-1}\sum_{j=1}^N z_j$ and the Lyapunov candidate is $V=|r|^2$. Its phase gradient is $\partial V/\partial\hat\theta_j=-(2/N)\operatorname{Im}(z_j\bar r)$. Continuous feedback $u_j=-k\partial V/\partial\hat\theta_j=(2k/N)\operatorname{Im}(z_j\bar r)$, with gain $k>0$, gives $\dot V=-k\|\nabla_\theta V\|^2$ when uncontrolled phase drift is absent or cancelled. In the neural implementation, apply $u_jJh_j$ with $J=[[0,-1],[1,0]]$, which rotates $h_j$ without changing its norm. At event time $t_\ell$, compute and store $u(t_\ell)$; between events use the held value $u(t)=u(t_\ell)$. Trigger the next update when $\|u(t)-u(t_\ell)\|_2\geq\delta$, where $\delta>0$. If $\|\dot u(t)\|_2\leq M$ during an interval, every inter-event time satisfies $t_{\ell+1}-t_\ell\geq\delta/M$, excluding Zeno behaviour. With held control, the ideal descent law becomes $\dot V=-k\|\nabla V\|^2+\nabla V^{\mathsf T}e$, where $e=u_{\mathrm{held}}-u_{\mathrm{exact}}$; thus a sufficiently small trigger tolerance gives practical rather than exact Lyapunov descent.

Implementation notes

First, integrate the mechanism into a continuous-time RNN or state-space model whose width is $2N$, interpreted as $N$ two-dimensional hidden channels $h_j=(x_j,y_j)$. After every solver step or discrete recurrent update, compute the pseudo-phases using atan2; for better numerical robustness, use normalized complex states z_j=(x_j+i*y_j)/(sqrt(x_j^2+y_j^2)+epsilon). Compute r, then u_exact[j]=(2*k/N)*imag(z_j*conj(r)). Initialize u_hold to zero. At each step, if norm(u_exact-u_hold) >= delta, replace u_hold with u_exact and record an event; otherwise retain the old value. Evolve each pair with h_j <- h_j + dt*(F_j(h,x;theta)+u_hold[j]*(J@h_j)), where J=[ [0,-1], [1,0] ]. The paper supplies the order-parameter Lyapunov mechanism, feedback gradient, trigger condition, and dwell-time argument. Estimate the bound $M$ empirically from a high percentile of finite differences of u_exact; tune $k$ and $delta$ on a validation set. Run a cheap experiment with a 32-channel continuous-time RNN on sequential MNIST, comparing no control, continuous control, and event-triggered control. Measure accuracy, $V(t)$, $\|\nabla V\|$, event rate, and minimum dwell time. The predicted signature is $\dot V<0$ under continuous control when intrinsic phase drift is negligible; event-triggered control should preserve the decay slope for small $\delta$, achieve event rate roughly proportional to $1/\delta$, and have minimum dwell time near or above $0.8\delta/\hat M$.

Verification

Mechanism confirmed, baseline not beaten

Stage 1 · Toy mechanism gate: Passed ✓

Stage 2 · Mechanism transferred to benchmark: Confirmed ✓

Stage 2 · Practical benchmark result: No measurable effect

Methodology: 8 paired seeds, permutation test (p<0.05) against a learning-rate-tuned baseline. How verification works

Stage 1 — Mechanism check agent confidence 8/10

Built a reproducible phase-controller verification and small recurrent classification experiment. The continuous Lyapunov law matched finite differences across gains k=0.25–2.0 with ratios 1.000002–1.000015, and event rates decreased approximately inversely with delta while dwell times tracked the empirical delta/M bound. The mechanism manifested, but it did not improve this classifier: baseline accuracy was 0.875 versus 0.852 event-triggered and 0.813 continuous.

Agent confidence: 8/10

Baseline: Toy initial Vdot at k=1: observed -0.004780887, predicted -0.004780850; RNN accuracy 0.875; zero events per sequence.
Idea: Event counts for delta [0.005, 0.02, 0.08, 0.2]: [128, 32, 6, 1]. Event RNN accuracy 0.8516 with 22.53 events per sequence; rotation norm error 2.61e-8.

Limitations: Only one seed and one tiny synthetic sequence-classification task were tested. Sequential MNIST, multi-seed statistics, tuning, FLOP or communication benchmarks, and larger recurrent/state-space models were not tested; the largest-delta run had only one event, so its minimum dwell time was unavailable.

How to run: CUDA_VISIBLE_DEVICES='' /home/maxwelhelp/main/bin/python3 experiment.py

Stage 2 — Benchmark vs. tuned baseline (arXiv:2608.14907) Sep 2, 2026

No significant difference vs. the tuned baseline (−-0.0%, p=1, wins 0 / 8 paired seeds).

Benchmark: Dynamics forecasting
Model: rnn_small
Paired seeds: 8
Baseline mean: 0.0001
Idea mean: 0
Effect (Δ): 0 (−-0.0%; negative = idea better)
Wins: 0 / 8 paired seeds
p-value: 1 (permutation test, 20 000 shuffles)
Smallest detectable effect: n/a
Mechanism: Confirmed ✓
Practical effect: No effect
Baseline tuning: swept over 0 configs

Limitations:

Only the structurally matched dynamics track was tested. The selected tolerance caused zero events, so nontrivial communication savings were not demonstrated. The local matched system used a 64-unit GRUCell rather than directly instantiating the bench rnn_small wrapper.

How to run: python3 bench_event.py

Verdict computed by deterministic test code from paired-seed statistics — not by the language model.

Stage 2 — Benchmark Sep 2, 2026 latest Failed ✗

Agent confidence: 10/10

Implemented event-triggered phase desynchronisation on the structurally matched dynamics track with 8 paired seeds and baseline/idea sweeps. The trained-model Lyapunov signature was quantitatively confirmed, but the selected controller generated zero events and matched the baseline MSE exactly; therefore there was no benchmark win.

Baseline: Mean test MSE 0.0001381416768708732 over 8 paired seeds; best lr=0.01, weight_decay=0.0.
Idea: Mean test MSE 0.0001381416768708732; paired delta=0.0; permutation p=1.0; best lr=0.01, k=0.5, delta=0.1; mean events per sequence=0.0.
Mechanism signature: Predicted exact one-step delta V=-0.0009295654599554837; observed=-0.0009223226156791497; ratio=0.9922083547761329; confirmed=true.

Benchmark result