arXiv research

Study of Hitchin map on specific Higgs bundles.

Classifies very stable Higgs bundles for complex groups.

Study very stable Higgs bundles on Riemann surfaces, linking to multiplicity and mirror symmetry.

This paper introduces Non-Autonomous Input-Output Stable Network(NAIS-Net), a very deep architecture where each stacked processing block is derived from a time-invariant non-autonomous dynamical system. Non-autonomy is implemented by skip connections from the block input to each of the unrolled processing stages and al…

We observe that several existing policy gradient methods (such as vanilla policy gradient, PPO, A2C) may suffer from overly large gradients when the current policy is close to deterministic (even in some very simple environments), leading to an unstable training process. To address this issue, we propose a new method, …

Study of parabolic Higgs bundles on curves with special fixed points.

New minimal surfaces grow area very quickly.

A machine learning model with approximate rotational symmetry is tested and found stable.

This paper extends homological stability results for configuration spaces of manifolds.

We focus on wind power modeling using machine learning techniques. We show on real data provided by the wind energy company Ma{ï}a Eolis, that parametric models, even following closely the physical equation relating wind production to wind speed are outperformed by intelligent learning algorithms. In particular, the CA…

Homotopy theory for $(2n+1)$-dimensional manifold triads with fixed boundary.

Recent work in reinforcement learning demonstrated that learning solely through self-play is not only possible, but could also result in novel strategies that humans never would have thought of. However, optimization methods cast as a game between two players require careful tuning to prevent suboptimal results. Hence,…

We consider expected performances based on max-stable random fields and we are interested in their derivatives with respect to the spatial dependence parameters of those fields. Max-stable fields, such as the Brown--Resnick and Smith fields, are very popular in spatial extremes. We focus on the two most popular unbiase…

Neural networks can approximate positive homogeneous functions, especially with multiple hidden layers.

We extend the Lyapunov-Schmidt analysis of outlying stable CMC spheres in the work of S. Brendle and the second-named author to the "far-off-center" regime and to include general Schwarzschild asymptotics. We obtain sharp existence and non-existence results for large stable CMC spheres that depend very delicately on th…

For each link L in S^3 and every quantum grading j, we construct a stable homotopy type X^j_o(L) whose cohomology recovers Ozsvath-Rasmussen-Szabo's odd Khovanov homology, H_i(X^j_o(L)) = Kh^{i,j}_o(L), following a construction of Lawson-Lipshitz-Sarkar of the even Khovanov stable homotopy type. Furthermore, the odd Kh…

CV inference can be invalid for relatively unstable model comparisons.

New stable minimal surfaces generalize classical Henneberg surface.

FGSM is more stable in adversarially robust transfer learning than PGD.

We develop methods to estimate lag and parameters for multiple stable autoregressive processes.

New robust estimator improves variable selection and coefficient estimation in linear regression with heavy-tailed errors and outliers.

We study the global behavior of (weakly) stable constant mean curvature hypersurfaces in general Riemannian manifolds. By using harmonic function theory, we prove some one-end theorems which are new even for constant mean curvature hypersurfaces in space forms. In particular, a complete oriented weakly stable minimal h…

Holomorphic connections found on Riemann surfaces with Fuchsian monodromy.

We present an approach to Gromov-Witten invariants that works on arbitrary (closed) symplectic manifolds. We avoid genericity arguments and take into account singular curves in the very formulation. The method is by first endowing mapping spaces from (prestable) algebraic curves into the symplectic manifold with the st…

Computes Steenrod squares on Khovanov homology for knots up to 11 crossings.

In this paper, we prove that the even solution of the mean field equation $Δu=λ(1-e^u)$ on $S^2$ must be axially symmetric when $4<λ\leq 8$. In particular, zero is the only even solution for $λ=6$. This implies the rigidity of Hawking mass for stable constant mean curvature(CMC) sphere with even symmetry.

To understand the empirical success of approximate MAP inference, recent work (Lang et al., 2018) has shown that some popular approximation algorithms perform very well when the input instance is stable. The simplest stability condition assumes that the MAP solution does not change at all when some of the pairwise pote…

Quaternion-Kähler manifolds' stability and rigidity of scalar curvature studied.

New algorithms improve policy evaluation in reinforcement learning.

Large learning rates work surprisingly well in standard parameterization, contrary to theory.

Dropout helps a stable network learn new tasks without forgetting old ones.

A model of open economics composed of producers and speculators is investigated by numerical simulations. The capital flows from the environment to the producers and from them to the speculators. The price fluctuations are suppressed by the speculators. When the aggressivity of the speculators grows, there is a transit…

Model-free deep reinforcement learning (RL) algorithms have been demonstrated on a range of challenging decision making and control tasks. However, these methods typically suffer from two major challenges: very high sample complexity and brittle convergence properties, which necessitate meticulous hyperparameter tuning…

Paper proves every stable 4-sphere has a unique diffeomorphism class.

Default-ERM shortcut learning persists even without additional information.

New LP method recovers MAP solution from noisy stable instances.

The paper explores symmetry in solutions of semilinear PDEs on Riemannian domains.

The paper shows how to stabilize off-policy reinforcement learning using specific state representations.

We propose a robust and stable lattice method which permits to obtain very accurate American option prices in presence of CIR stochastic interest rate without any numerical restriction on its parameters. Numerical results show the reliability and the accuracy of the proposed method.

Stablecoins are unstable, but some are more stable than others.

Cooper-Manning and Louder gave examples of maps of surface groups to PSL(2,C) which are not injective, but are incompressible (i.e. no simple loop is in the kernel). We construct more examples with very simple certificates for their incompressibility arising from the theory of stable commutator length.

A faster, more stable method for optimizing topological functions.

In this paper we introduce the notion of cofrontal mappings, as the dual objects to frontal mappings, and study their basic local and global properties. Cofrontals are very special mappings and far from generic nor stable except for the case of submersions. It is observed that any smooth mapping can be $C^0$-approximat…

New constructions show stable geodesics and figure-eights in convex hypersurfaces.

This paper introduces even triangulations of n-dimensional pseudo-manifolds and links their combinatorics to the topology of the pseudo-manifolds. This is done via normal hypersurface theory and the study of certain symmetric representation. In dimension 3, necessary and sufficient conditions for the existence of even …

DISTANA predicts and denoises spatial wave dynamics.

Learning optimal dictionaries for sparse coding has exposed characteristic sparse features of many natural signals. However, universal guarantees of the stability of such features in the presence of noise are lacking. Here, we provide very general conditions guaranteeing when dictionaries yielding the sparsest encoding…

Proposes a new model for clustering with heavier tails.