ML: Uncertainty

Machine-learning ideas tagged Uncertainty in the ML taxonomy of the Math2NN corpus.

2 ideas found

✓✓ Beats tuned baseline 2026

Conjugate Bayesian Latent Dynamics Head

Replace the final nonlinear transition network of a latent world model with a linear Koopman-style transition whose coefficients have a Matrix Normal-Inverse Wishart prior. Meta-learn the prior across tasks, then adapt only closed-form sufficient statistics from a few recent transitions; this should be more data-efficient and uncertainty-aware than gradient fine-tuning under distribution shift.

Useful7/10

Difficulty5/10

Novelty6/10

Dynamics Stability Sample efficiency Generalization

Paper: MetaKoopman: Bayesian Meta-Learning of Koopman Operators for Modeling Structured Dynamics under Distribution Shifts arXiv:2607.26345

Unverified 2026

Log-Euclidean SPD prediction head

Replace a neural network head that predicts a covariance or other SPD matrix entrywise with regression in the matrix-log domain. The network predicts a symmetric matrix in unconstrained Euclidean coordinates, the matrix exponential guarantees an SPD output, and training can use intrinsic log-Euclidean or affine-invariant errors rather than Frobenius error on raw entries.

Useful6/10

Difficulty4/10

Novelty5/10

Geometry Stability Accuracy Generalization

Paper: Geometric turbulence: a geodesic-regression crisis indicator for equity covariance dynamics, with evidence from African markets arXiv:2608.15205