Linear Speedup In Saddle-Point Escape For Decentralized Non-Convex Optimization

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

A traditional machine learning pipeline involves collecting massive amounts of data centrally on a server and training models to fit the data. However, increasing concerns about the privacy and security of user's data, combined with the sheer growth in the ...

This work presents the study of suspended Y-cut Lithium Niobate shear bulk acoustic resonators for wide band filter applications in the frequency range of 3.5-4.5GHz. The resonators consist of a Lithium Niobate film with Aluminum top interdigitated electro ...

In this work, we propose an analytical model for the intrinsic transcapacitances in ultra-thin gate-all-around junctionless nanowire field effect transistors in the presence of confined energy states of electrons. The validity of the developed model is con ...

This paper presents an adaptive combination strategy for distributed learning over diffusion networks. Since learning relies on the collaborative processing of the stochastic information at the dispersed agents, the overall performance can be improved by d ...

This work deals with a control strategy to regulate homogeneous reaction systems in reactors equipped with a cooling jacket. The reaction system is expressed in terms of vessel extents and viewed as a single-input (the inlet coolant temperature), two-outpu ...

The research community has been making significant progress in hardware implementation, numerical computing and algorithm development for optimization-based control. However, there are two key challenges that still have to be overcome for optimization-base ...

Over the past two decades, the thermo-hydro-mechanical behavior of earth-contact structures such as piles, walls, slabs, and tunnels, which simultaneously provide structural support and energy supply, has been studied through various field experiments and ...

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 ...

We consider the problem of compressing an information source when a correlated one is available as side information only at the decoder side, which is a special case of the distributed source coding problem in information theory. In particular, we consider ...

We present a computational inverse design method for a new class of surface-based inflatable structure. Our deployable structures are fabricated by fusing together two layers of inextensible sheet material along carefully selected curves. The fusing curves ...

A programmable patient-customized epileptic seizure detector is proposed in this paper to enable neurologists and patients to have constructive interactions with the implantable medical device. The programmability feature is enabled by designing a low-powe ...

Federated learning is a useful framework for centralized learning from distributed data under practical considerations of heterogeneity, asynchrony, and privacy. Federated architectures are frequently deployed in deep learning settings, which generally giv ...

Non-convex constrained optimization problems have become a powerful framework for modeling a wide range of machine learning problems, with applications in k-means clustering, large- scale semidefinite programs (SDPs), and various other tasks. As the perfor ...

Rapid advances in data collection and processing capabilities have allowed for the use of increasingly complex models that give rise to nonconvex optimization problems. These formulations, however, can be arbitrarily difficult to solve in general, in the s ...

Background: Although digital mobility outcomes (DMOs) can be readily calculated from real-world data collected with wearable devices and ad-hoc algorithms, technical validation is still required. The aim of this paper is to comparatively assess and validat ...

Despite the widespread use of gradient-based algorithms for optimizing high-dimensional non-convex functions, understanding their ability of finding good minima instead of being trapped in spurious ones remains to a large extent an open problem. Here we fo ...

Current procedures for the rapid inspection of buildings and infrastructure are subjective, time-consuming, and cumbersome to document, necessitating new technologies to automate the process and eliminate these shortcomings. Fortunately, recent development ...

Binary dielectric metasurfaces are arrays of sub-wavelength structures that act as a thin layer of artificial material. They are generally lossless and relatively simple to fabricate since only a single structuring step is required. By carefully designing ...

Stable and uniform beams with low divergence are required in particle accelerators; therefore, beyond the accelerated current, measuring the beam current spatial uniformity and stability over time is necessary to assess the beam performance, since these pa ...

We present a strikingly simple proof that two rules are sufficient to automate gradient descent: 1) don’t increase the stepsize too fast and 2) don’t overstep the local curvature. No need for functional values, no line search, no information about the func ...

Linear Speedup In Saddle-Point Escape For Decentralized Non-Convex Optimization

Chat with Graph Search

Optimization methods for collaborative learning

Toward Band n78 Shear Bulk Acoustic Resonators Using Crystalline Y-Cut Lithium Niobate Films With Spurious Suppression

Transcapacitances Modeling in ultra-thin gate-all-around junctionless nanowire FETs, including 2D quantum confinement

Gramian-Based Adaptive Combination Policies For Diffusion Learning Over Networks

Output regulation via trajectory tracking for chemical reactors with jacket dynamics

Optimization Methods for Control: From Embedded Programmable Hardware to Data-Driven Process Optimization

Thermo-hydro-mechanical behavior of energy barrettes: Field experiments and numerical simulations

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

Neural Distributed Image Compression with Cross-Attention Feature Alignment

Computational Inverse Design of Surface-based Inflatables

Programmable Seizure Detector Using a 32-bit RISC Processor for Implantable Medical Devices

Second-Order Guarantees In Federated Learning

Augmented Lagrangian Methods for Provable and Scalable Machine Learning

Second-order guarantees in centralized, federated and decentralized nonconvex optimization

Assessing real-world gait with digital technology? Validation, insights and recommendations from the Mobilise-D consortium

Complex Dynamics in Simple Neural Networks: Understanding Gradient Flow in Phase Retrieval

Damage-augmented digital twins towards the automated inspection of buildings

On design strategies for binary dielectric metasurfaces based on the Fourier modal method

Design and Development of a Diagnostic System for a Non-Intercepting Direct Measure of the SPIDER Ion Source Beamlet Current

Adaptive Gradient Descent without Descent