# Periodic-Orbit Continuation for Recurrent Inference

- ID: 2655
- Canonical URL: https://synthcore.org/idea/2655/periodic-orbit-continuation-for-recurrent-inference
- API JSON: https://synthcore.org/api/idea/2655.json
- API Markdown: https://synthcore.org/api/idea/2655.md
- Verification status: failed_benchmark
- Source: [arXiv:2608.21256](https://arxiv.org/abs/2608.21256)
- Category: dynamics
- Solves: stability, speedup
- ML areas: rnn, ssm, inference-speedup
- Math tags: dynamical-systems, bifurcations, linear-algebra, numerical-analysis
- Ratings: usefulness 7/10; difficulty 7/10; novelty 7/10

## Idea description

For a recurrent or implicit neural model driven by periodic inputs, solve for a periodic hidden-state orbit and continue that orbit as input amplitude or frequency changes. This replaces repeated cold starts from zero and should preserve convergence near parameter ranges where cold starts fail.

## Mathematical statement

Let z_{k+1}=G_k(z_k;p) be a recurrent or implicit neural update with period T and forcing parameter p. Define the period residual R(z,p)=Phi_T(z;p)-z, where Phi_T is the T-step state map. A periodic state satisfies R(z,p)=0. Differentiation along a regular branch gives (D_z Phi_T-I) dz/dp = -D_p Phi_T. Continuation becomes ill-conditioned when D_z Phi_T-I is singular. Since M=D_z Phi_T, this is the same as a Floquet multiplier +1. The smallest singular value sigma_min(I-M) therefore measures proximity to the cyclic-fold boundary.

## Key formulas

- $$R(z,p)=\Phi_T(z;p)-z=0$$
- $$(D_z\Phi_T-I)\frac{dz}{dp}=-D_p\Phi_T$$
- $$\sigma_{\min}(I-M)\to 0,\qquad M=D_z\Phi_T$$
- $$R(z,p)=0,\qquad c_z^{\top}(z-z_{\mathrm{pred}})+c_p(p-p_{\mathrm{pred}})-\Delta s=0$$

## Implementation notes

Apply the method to a tanh RNN, an SSM, or a DEQ layer receiving a periodic input. The integration point is inference-time state initialization. At an easy amplitude p_0, solve R(z,p_0)=0 by ordinary fixed-point iteration and store the full period orbit. For the next parameter value, predict the orbit using a secant step from the last two converged solutions, then correct it with Newton or quasi-Newton iterations on the period residual together with the pseudo-arclength constraint. Compute JVPs and VJPs through the unrolled period map, and estimate sigma_min(I-M) with randomized SVD or Lanczos rather than forming M explicitly. The paper contributes the fixed-point residual, continuation equation, and singularity criterion; the neural Jacobians and orbit residuals are estimated empirically. The first cheap experiment should use a small tanh RNN on a synthetic sinusoidal sequence and sweep input amplitude. Compare zero-state fixed-point solves against continuation warm starts, measuring iteration count, residual, and prediction error. The quantitative prediction is that cold starts lose convergence near a fold while continuation follows the orbit farther, and sigma_min(I-M) decreases toward zero as the leading multiplier approaches +1.

## Verification

- Status: failed_benchmark
- Mechanism evidence: yes
- Mechanism confirmed: no
- Practical verdict: inconclusive
- Verdict: Built a readable scalar tanh-RNN periodic-orbit continuation MVP with fixed-point solving, pseudo-arclength correction, Floquet multiplier and singular-gap diagnostics, and a cold-start comparison. The checks confirmed the analytic fold at p=-0.207546 with multiplier +1, Jacobian error 2.2e-11, square-root gap scaling with observed exponent 0.456 versus predicted 0.5, and nearly constant iteration-times-gap products. Continuation tracked the connected orbit to within 3.8e-5 of the fold, turned around, and returned along the branch with residual below 7.7e-11; this demonstrates the mechanism, although cold starts did not diverge and instead selected a different stable branch.

### Mechanism check

- Verdict: Built a readable scalar tanh-RNN periodic-orbit continuation MVP with fixed-point solving, pseudo-arclength correction, Floquet multiplier and singular-gap diagnostics, and a cold-start comparison. The checks confirmed the analytic fold at p=-0.207546 with multiplier +1, Jacobian error 2.2e-11, square-root gap scaling with observed exponent 0.456 versus predicted 0.5, and nearly constant iteration-times-gap products. Continuation tracked the connected orbit to within 3.8e-5 of the fold, turned around, and returned along the branch with residual below 7.7e-11; this demonstrates the mechanism, although cold starts did not diverge and instead selected a different stable branch.
- Confidence: 8/10
- Limitations: The experiment uses a scalar T=1 tanh recurrence rather than a multidimensional RNN with genuinely time-periodic forcing. It does not benchmark wall-clock inference speed, GPU execution, prediction error, randomized SVD, or a case where cold fixed-point iteration truly fails; the continuation trace approaches the fold but does not numerically cross to p below it because the chosen step size and tolerance stop 3.8e-5 short.

### Practical benchmark

- Paired seeds: 8
- Baseline mean: 0
- Idea mean: 0
- p-value: 0.055
- Paired wins: 6/8
- Benchmark verdict: no significant win

## Artifacts

- [bench_experiment.py](https://synthcore.org/code/1015/bench_experiment.py)
- [bench_report.json](https://synthcore.org/code/1015/bench_report.json)
- [experiment.py](https://synthcore.org/code/1015/experiment.py)
- [report.md](https://synthcore.org/code/1015/report.md)
- [report_bench_2026-09-01T182108.md](https://synthcore.org/code/1015/report_bench_2026-09-01T182108.md)
- [results.json](https://synthcore.org/code/1015/results.json)
- [Download all files as ZIP](https://synthcore.org/download/1015)

## Disclaimer

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