arXiv research

The paper constructs infinitely many $G$-smoothings of a $G$-manifold.

We are interested in global properties of systems of left-invariant differential operators on compact Lie groups: regularity properties, properties on the closedness of the range and finite dimensionality of their cohomology spaces, when acting on various function spaces e.g. smooth, analytic and Gevrey. Extending the …

This work introduces the concept of tangent space regularization for neural-network models of dynamical systems. The tangent space to the dynamics function of many physical systems of interest in control applications exhibits useful properties, e.g., smoothness, motivating regularization of the model Jacobian along sys…

Imitation Learning describes the problem of recovering an expert policy from demonstrations. While inverse reinforcement learning approaches are known to be very sample-efficient in terms of expert demonstrations, they usually require problem-dependent reward functions or a (task-)specific reward-function regularizatio…

Novikov theorem extended to rational Pontryagin classes for cyclic group $C_4$.

This paper presents the beginnings of an automatic statistician, focusing on regression problems. Our system explores an open-ended space of statistical models to discover a good explanation of a data set, and then produces a detailed report with figures and natural-language text. Our approach treats unknown regression…

Recent developments in linear system identification have proposed the use of non-parameteric methods, relying on regularization strategies, to handle the so-called bias/variance trade-off. This paper introduces an impulse response estimator which relies on an $\ell_2$-type regularization including a rank-penalty derive…

Interpretable deep learning model for insurance pricing.

Study on braid monodromy of Lefschetz fibrations, proving infinite index subgroup.

New algorithms reduce regret in online convex optimization with heavy-tailed gradients.

Deep neural networks are often trained in the over-parametrized regime (i.e. with far more parameters than training examples), and understanding why the training converges to solutions that generalize remains an open problem. Several studies have highlighted the fact that the training procedure, i.e. mini-batch Stochas…

Post-processing predictors reduces calibration errors for decision-making.

The paper explores smooth equivariant rigidity and finds infinitely many exotic smooth structures.

Algorithm learns without knowing distribution, reducing error.

BMM algorithm improves convergence for nonconvex optimization problems.

Epoch-GDA achieves optimal convergence rate for SCSC min-max problems.

Deep neural networks perform poorly on smooth functions despite strong approximation theory.