# Critical-Set Cone Monitor

- ID: 2541
- Canonical URL: https://synthcore.org/idea/2541/critical-set-cone-monitor
- API JSON: https://synthcore.org/api/idea/2541.json
- API Markdown: https://synthcore.org/api/idea/2541.md
- Verification status: unverified
- Source: [arXiv:2608.18761](https://arxiv.org/abs/2608.18761)
- Category: dynamics
- Solves: stability, accuracy
- ML areas: rnn, ssm, world-model, training-dynamics, regularization
- Math tags: dynamical-systems, bifurcations, geometry, linear-algebra
- Ratings: usefulness 6/10; difficulty 7/10; novelty 8/10

## Idea description

Add a Jacobian cone-field regularizer to recurrent dynamics so that tangent directions expand and remain aligned with an unstable cone outside a designated critical neighborhood. The network is not forced to be uniformly expanding: the regularizer is disabled near the critical set, allowing controlled bifurcation-like behavior while exposing where long-horizon sensitivity changes.

## Mathematical statement

For a differentiable map f and Jacobian Df_x, let C^u_(x,rho) and C^s_(x,rho) be unstable and stable cones with aperture parameter rho in (0,1]. Let C_delta be a critical neighborhood of radius delta, and let K be a compact state region. The paper's hyperbolicity mechanism is strict cone invariance and expansion while an orbit avoids C_delta: Df_x^i(C^u_(x,rho)) is contained in the interior of C^u_(f^i(x),rho), and ||Df_x^i v|| >= c0 theta exp(c1 i) ||v|| for constants c0,c1 > 0 and distortion factor theta. In a neural model, use a unit tangent q_x and impose one-step expansion ||Df_x q_x|| >= exp(gamma), plus alignment of the normalized vector Df_x q_x with q_(f(x)). The critical set prevents imposing this condition where the theorem does not provide it.

## Key formulas

- Df_x^i(C^u_(x,rho)) is contained in the interior of C^u_(f^i(x),rho)
- ||Df_x^i v|| >= c0 theta exp(c1 i) ||v|| for v in C^u_(x,rho)
- S(delta) = sup of |(Df_x)_1| over x in K minus C_delta
- L_cone = mean over x outside C_delta of [softplus(gamma - log(||J_x q_x|| + 1e-8)) + beta ReLU(angle(J_x q_x, q_(f(x))) - rho)]

## Implementation notes

1. Exact integration point: use a recurrent transition h_(t+1) = f_theta(h_t) and apply the regularizer only when h_t is outside C_delta = {h: ||h - h_c||_2 < delta}. Let q_h be a unit tangent direction, initialized randomly and updated by tangent propagation. 2. Pseudocode: sample h from replay trajectories; compute the Jacobian-vector product v = J_h q_h using automatic differentiation; set q_new = v divided by (||v|| + 1e-8); calculate expansion_loss = softplus(gamma - log(||v|| + 1e-8)); calculate angle_loss from the angle between q_new and the stored tangent direction at f_theta(h); if ||h - h_c|| >= delta, add expansion_loss + beta times angle_loss to the task loss; otherwise add zero; update q_h to q_new after detaching it from the graph. 3. Taken from the paper: the critical-set exclusion, invariant-cone condition, and exponential tangent expansion. Estimated empirically: Jacobian-vector products, cone angles, finite-time Lyapunov growth, and the critical-set center h_c. 4. First cheap experiment: train a 32-unit RNN on Lorenz or Mackey-Glass prediction, comparing task-only training against task plus cone regularization. Sweep gamma and delta, and measure g_i = (1/i) log(||J_(h_(i-1)) ... J_(h_0) q_0|| / ||q_0||). The predicted signature is a growth plateau near gamma for trajectories avoiding C_delta, followed by a reproducible drop or sign change after critical-set entry. Increasing delta should shorten the certified expansion interval, while the plateau level should remain approximately unchanged.

## Disclaimer

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