
On EnergyBased Models with Overparametrized Shallow Neural Networks
Energybased models (EBMs) are a simple yet powerful framework for gener...
read it

Symmetry Breaking in Symmetric Tensor Decomposition
In this note, we consider the optimization problem associated with compu...
read it

Depth separation beyond radial functions
Highdimensional depth separation results for neural networks show that ...
read it

SelfSupervised Equivariant Scene Synthesis from Video
We propose a selfsupervised framework to learn scene representations fr...
read it

Learned Equivariant Rendering without Transformation Supervision
We propose a selfsupervised framework to learn scene representations fr...
read it

On Graph Neural Networks versus GraphAugmented MLPs
From the perspective of expressive power, this work compares multilayer...
read it

KernelBased Smoothness Analysis of Residual Networks
A major factor in the success of deep neural networks is the use of soph...
read it

A Dynamical Central Limit Theorem for Shallow Neural Networks
Recent theoretical work has characterized the dynamics of wide shallow n...
read it

A Functional Perspective on Learning Symmetric Functions with Neural Networks
Symmetric functions, which take as input an unordered, fixedsize set, a...
read it

Depth separation for reduced deep networks in nonlinear model reduction: Distilling shock waves in nonlinear hyperbolic problems
Classical reduced models are lowrank approximations using a fixed basis...
read it

InDistribution Interpretability for Challenging Modalities
It is widely recognized that the predictions of deep neural networks are...
read it

Overfitting and Optimization in Offline Policy Learning
We consider the task of policy learning from an offline dataset generate...
read it

Neural Splines: Fitting 3D Surfaces with InfinitelyWide Neural Networks
We present Neural Splines, a technique for 3D surface reconstruction tha...
read it

On Sparsity in Overparametrised Shallow ReLU Networks
The analysis of neural network training beyond their linearization regim...
read it

IDEAL: Inexact DEcentralized Accelerated Augmented Lagrangian Method
We introduce a framework for designing primal methods under the decentra...
read it

Continuous LWE
We introduce a continuous analogue of the Learning with Errors (LWE) pro...
read it

A PermutationEquivariant Neural Network Architecture For Auction Design
Designing an incentive compatible auction that maximizes expected revenu...
read it

Provably Efficient ThirdPerson Imitation from Offline Observation
Domain adaptation in imitation learning represents an essential step tow...
read it

A meanfield analysis of twoplayer zerosum games
Finding Nash equilibria in twoplayer zerosum continuous games is a cen...
read it

Can graph neural networks count substructures?
The ability to detect and count certain substructures in graphs is impor...
read it

Probing the State of the Art: A Critical Look at Visual Representation Evaluation
Selfsupervised research improved greatly over the past half decade, wit...
read it

Stability of Graph Neural Networks to Relative Perturbations
Graph neural networks (GNNs), consisting of a cascade of layers applying...
read it

Pure and Spurious Critical Points: a Geometric Study of Linear Networks
The critical locus of the loss function of a neural network is determine...
read it

Gradient Dynamics of Shallow Univariate ReLU Networks
We present a theoretical and empirical study of the gradient dynamics of...
read it

Finding the Needle in the Haystack with Convolutions: on the benefits of architectural bias
Despite the phenomenal success of deep neural networks in a broad range ...
read it

Stability of Graph Scattering Transforms
Scattering transforms are nontrainable deep convolutional architectures...
read it

On the equivalence between graph isomorphism testing and function approximation with GNNs
Graph neural networks (GNNs) have achieved lots of success on graphstru...
read it

Extragradient with player sampling for provable fast convergence in nplayer games
Datadriven model training is increasingly relying on finding Nash equil...
read it

On the Expressive Power of Deep Polynomial Neural Networks
We study deep neural networks with polynomial activations, particularly ...
read it

On the Expected Dynamics of Nonlinear TD Learning
While there are convergence guarantees for temporal difference (TD) lear...
read it

Stability Properties of Graph Neural Networks
Data stemming from networks exhibit an irregular support, whereby each d...
read it

Advancing GraphSAGE with A DataDriven Node Sampling
As an efficient and scalable graph neural network, GraphSAGE has enabled...
read it

Global convergence of neuron birthdeath dynamics
Neural networks with a large number of parameters admit a meanfield des...
read it

Kymatio: Scattering Transforms in Python
The wavelet scattering transform is an invariant signal representation s...
read it

Deep Geometric Prior for Surface Reconstruction
The reconstruction of a discrete surface from a point cloud is a fundame...
read it

Pommerman: A MultiAgent Playground
We present Pommerman, a multiagent environment based on the classic con...
read it

Graph Neural Networks for IceCube Signal Classification
Tasks involving the analysis of geometric (graph and manifoldstructure...
read it

Planning with Arithmetic and Geometric Attributes
A desirable property of an intelligent agent is its ability to understan...
read it

Backplay: "Man muss immer umkehren"
A longstanding problem in model free reinforcement learning (RL) is tha...
read it

Diffusion Scattering Transforms on Graphs
Stability is a key aspect of data analysis. In many applications, the na...
read it

Neural Networks with Finite Intrinsic Dimension have no Spurious Valleys
Neural networks provide a rich class of highdimensional, nonconvex opt...
read it

Multiscale Sparse Microcanonical Models
We study density estimation of stationary processes defined over an infi...
read it

Mathematics of Deep Learning
Recently there has been a dramatic increase in the performance of recogn...
read it

FewShot Learning with Graph Neural Networks
We propose to study the problem of fewshot learning with the prism of i...
read it

A Note on Learning Algorithms for Quadratic Assignment with Graph Neural Networks
Many inverse problems are formulated as optimization problems over certa...
read it

Understanding the Learned Iterative Soft Thresholding Algorithm with matrix factorization
Sparse coding is a core building block in many data analysis and machine...
read it

Surface Networks
We study datadriven representations for threedimensional triangle mesh...
read it

Community Detection with Graph Neural Networks
We study datadriven methods for community detection in graphs. This est...
read it

Geometric deep learning: going beyond Euclidean data
Many scientific fields study data with an underlying structure that is a...
read it

Divide and Conquer Networks
We consider the learning of algorithmic tasks by mere observation of inp...
read it