Publications related to Direct sum

Adaptive Stochastic Variance Reduction for Non-convex Finite-Sum Minimization

Volkan Cevher, Kimon Antonakopoulos, Efstratios Panteleimon Skoulakis, Leello Tadesse Dadi, Ali Kavis

We propose an adaptive variance-reduction method, called AdaSpider, for minimization of L-smooth, non-convex functions with a finite-sum structure. In essence, AdaSpider combines an AdaGrad-inspired [Duchi et al., 2011, McMahan & Streeter, 2010], but a fai ...

2022

Stochastic distributed learning with gradient quantization and double-variance reduction

Sebastian Urban Stich, Konstantin Mishchenko

We consider distributed optimization over several devices, each sending incremental model updates to a central server. This setting is considered, for instance, in federated learning. Various schemes have been designed to compress the model updates in orde ...

TAYLOR & FRANCIS LTD2022

Collapse-invariant properties of spaces equipped with signals or directions

Stefania Ebli

Collapsing cell complexes was first introduced in the 1930's as a way to deform a space into a topological-equivalent subspace with a sequence of elementary moves. Recently, discrete Morse theory techniques provided an efficient way to construct deformatio ...

EPFL2022

Idempotents in the endomorphism algebra of a finite lattice

Jacques Thévenaz, Serge Bouc

We give a direct construction of a specific central idempotent in the endomorphism algebra of a finite lattice T. This idempotent is associated with all possible sublattices of T which are totally ordered. A generalization is considered in a conjectural fa ...

2022

Intermolecular Pd(0)-Catalyzed Atropo-enantioselective C-H Arylation of Heteroarenes and Other Investigations into the Enantioselective Activation of Challenging C(sp3)-H Bonds

Qui-Hien Nguyen

The direct construction of stereogenic axes by enantioselective C-H arylation has been regarded as a very challenging transformation. Very few examples have been reported, despite the indisputable straight-forwardness of the methodology. This is due to the ...

EPFL2021

A generalized theory of separable effects in competing event settings

Mats Julius Stensrud

In competing event settings, a counterfactual contrast of cause-specific cumulative incidences quantifies the total causal effect of a treatment on the event of interest. However, effects of treatment on the competing event may indirectly contribute to thi ...

2021

The effects of Prandtl number on the nonlinear dynamics of Kelvin-Helmholtz instability in two dimensions

Jeremy Peter Parker

It is known that the pitchfork bifurcation of Kelvin-Helmholtz instability occurring at minimum gradient Richardson number Ri(m) similar or equal to 1/4 in viscous stratified shear flows can be subcritical or supercritical depending on the value of the Pra ...

Cambridge University Press2021

Development of remote C-H bond functionalization of redox-active alcohol derivatives and oxidative transformations under photoredox catalysis

Alexandre Julien Florent Leclair

The development of radical-based remote C(sp3)-H functionalizations of redox-active alcohol derivatives, oxidative cycloaddition and rearrangement transformations under visible light photocatalysis constitute the basis of this thesis. The first chapter des ...

EPFL2021

Poly-NL: Linear Complexity Non-local Layers With 3rd Order Polynomials

Grigorios Chrysos, Filippos Kokkinos

Spatial self-attention layers, in the form of Non-Local blocks, introduce long-range dependencies in Convolutional Neural Networks by computing pairwise similarities among all possible positions. Such pairwise functions underpin the effectiveness of non-lo ...

IEEE2021

Convex optimization in sums of Banach spaces

Michaël Unser, Shayan Aziznejad

We characterize the solution of a broad class of convex optimization problems that address the reconstruction of a function from a finite number of linear measurements. The underlying hypothesis is that the solution is decomposable as a finite sum of compo ...

2021

Time-Synchronization Attack Detection in Unbalanced Three-Phase Systems

Jean-Yves Le Boudec, Mario Paolone, Marguerite Marie Nathalie Delcourt, Gyorgy Miklos Dan

Phasor measurement units (PMU) rely on an accurate time-synchronization to phase-align the phasors and timestamp the voltage and current phasor measurements. Among the symmetrical components computed from the phasors in three-phase systems, the standard pr ...

2021

Adaptation in Stochastic Algorithms: From Nonsmooth Optimization to Min-Max Problems and Beyond

Ahmet Alacaoglu

Stochastic gradient descent (SGD) and randomized coordinate descent (RCD) are two of the workhorses for training modern automated decision systems. Intriguingly, convergence properties of these methods are not well-established as we move away from the spec ...

EPFL2021

A Catalyst Framework for Minimax Optimization

Negar Kiyavash

We introduce a generic two-loop scheme for smooth minimax optimization with strongly-convex-concave objectives. Our approach applies the accelerated proximal point framework (or Catalyst) to the associated dual problem and takes full advantage of existing ...

NEURAL INFORMATION PROCESSING SYSTEMS (NIPS)2020

Global Convergence and Variance Reduction for a Class of Nonconvex-Nonconcave Minimax Problems

Negar Kiyavash

Nonconvex minimax problems appear frequently in emerging machine learning applications, such as generative adversarial networks and adversarial learning. Simple algorithms such as the gradient descent ascent (GDA) are the common practice for solving these ...

2020

Laser-induced densification of fused silica using spatially overlapping sub-30fs pulses

Yves Bellouard, Farhang Nabiei, Ya Cheng

Over the past several years, non-ablative femtosecond laser exposure with spatially overlapping (i.e., spatially cumulative) pulses has emerged as a key process in three-dimensional writing of patterns in bulk of dielectric substrates. When temporally non- ...

2020

Momentum space CFT correlators for Hamiltonian truncation

Matthew Thomas Walters

We consider Lorentzian CFT Wightman functions in momentum space. In particular, we derive a set of reference formulas for computing two- and three-point functions, restricting our attention to three-point functions where the middle operator (corresponding ...

Springer Nature2020

Loops in AdS: from the spectral representation to position space

Din Carmi

We compute a family of scalar loop diagrams in AdS. We use the spectral representation to derive various bulk vertex/propagator identities, and these identities enable to reduce certain loop bubble diagrams to lower loop diagrams, and often to tree- level ...

Springer Nature2020

The covering radius and a discrete surface area for non-hollow simplices

Matthias Schymura

We explore upper bounds on the covering radius of non-hollow lattice polytopes. In particular, we conjecture a general upper bound of d/2 in dimension d, achieved by the "standard terminal simplices" and direct sums of them. We prove this conjecture up to ...

2022

Scanning Probe Microscopy Characterization of Organic Self-Assembled Monolayers

Evangelia-Nefeli Athanasopoulou

As technology advances, surface coatings become more and more important to assure materials performances. In recent years, molecular coatings have found larger acceptance and uses. Among them, self-assembled monolayers (SAMs) are attractive because they ar ...

EPFL2018

Combinatorial presentation of multidimensional persistent homology

Martina Scolamiero

A multifiltration is a functor indexed by Nr that maps any morphism to a monomorphism. The goal of this paper is to describe in an explicit and combinatorial way the natural Nr-graded R[x(1),...x(r)]-module structure on the homology of a multifiltration of ...

Elsevier2017