Research ideas

Every idea extracted from recent arXiv mathematics papers — verified and unverified. Click an idea to open its full card; badges show the empirical verdict.

2 ideas found

Status: Beats tuned baseline ML: Mlp Math: Control theory Clear all

✓✓ Beats tuned baseline 2026

Power-Balanced Modular Neural Block

Represent each neural module as a Hamiltonian storage system and connect modules through a state-dependent skew or Dirac interconnection instead of arbitrary residual additions. The coupling may change with the hidden state, but its internal power contribution cancels exactly, so total stored energy is controlled only by external inputs and explicitly added dissipation.

Useful8/10

Difficulty5/10

Novelty5/10

Architecture Stability Generalization Accuracy

Paper: Port-Hamiltonian modelling of coupled rigid/flexible multibody systems arXiv:2608.05143

✓✓ Beats tuned baseline 2026

Uniform-Certificate Bayesian Feature Head

Replace the final layer of a neural predictor with Bayesian linear regression over deterministic trigonometric features, retaining a computable posterior variance and a high-probability confidence envelope over the full bounded input domain. Use this envelope to reject unsafe actions, downweight uncertain training targets, or restrict optimizer updates in regions where the network is extrapolating.

Useful7/10

Difficulty5/10

Novelty6/10

Regularization Stability Generalization Scalability

Paper: Scalable Gaussian Process Regression via Deterministic Trigonometric Features: Uniform Bounds for Safe Model Predictive Control arXiv:2608.16415