Concept

Overfitting

Related publications (93)

Learning Multi-Label Aerial Image Classification Under Label Noise: A Regularization Approach Using Word Embeddings

Training deep neural networks requires well-annotated datasets. However, real world datasets are often noisy, especially in a multi-label scenario, i.e. where each data point can be attributed to more than one class. To this end, we propose a regularizatio ...

IEEE2020

Simultaneous autoregressive models for spatial extremes

Emeric Rolland Georges Thibaud

Motivated by the widespread use of large gridded data sets in the atmospheric sciences, we propose a new model for extremes of areal data that is inspired by the simultaneous autoregressive (SAR) model in classical spatial statistics. Our extreme SAR model ...

WILEY2020

A Novel Spatial-Angular Domain Regularisation Approach for Restoration of Diffusion MRI

Jean-Philippe Thiran, Meritxell Bach Cuadra, Alessandro Daducci, Alessandro Mella

In this paper we tackle the problem of regularisation for inverse problems in single shell diffusion weighted image restoration. Our aim is to recover a high-resolution and denoised DWI signal, prior to any model fitting. The main contribution of our metho ...

SPRINGER-VERLAG BERLIN2019

Experimental Verification of Sum-Of-Squares-Based Controller Tuning Technique with Extension to Parallel Multimodel Uncertainty Processing

Colin Neil Jones, Ivan Pejcic

Many control schemes require significant tuning effort to achieve desired performance targets, causing a need for general tools that can perform controller tuning in an automated fashion. This paper represents a continuation of the previous work in which a ...

IEEE2019

Weak teachers: assisted specification of discrete choice models using ensemble learning algorithms

Timothy Michael Hillel

Discrete Choice Models (DCMs) have a distinct advantage over Machine Learning (ML) classification algorithms, in that they employ a highly interpretable linear structure. However, a key drawback of DCMs compared to ML is the need to specify the utility fun ...

2019

Exploiting semi-supervised training through a dropout regularization in end-to-end speech recognition

Petr Motlicek, Subhadeep Dey

In this paper, we explore various approaches for semi- supervised learning in an end-to-end automatic speech recog- nition (ASR) framework. The first step in our approach in- volves training a seed model on the limited amount of labelled data. Additional u ...

2019

Reconstructing the Drizzle Mode of the Raindrop Size Distribution Using Double-Moment Normalization

Alexis Berne, Timothy Hugh Raupach

Commonly used disdrometers tend not to accurately measure concentrations of very small drops in the raindrop size distribution (DSD), either through truncation of the DSD at the small-drop end or because of large uncertainties on these measurements. Recent ...

AMER METEOROLOGICAL SOC2019

Non-intrusive reduced order modeling of nonlinear problems using neural networks

Jan Sickmann Hesthaven, Stefano Ubbiali

We develop a non-intrusive reduced basis (RB) method for parametrized steady-state partial differential equations (PDEs). The method extracts a reduced basis from a collection of high-fidelity solutions via a proper orthogonal decomposition (POD) and emplo ...

2018

Cross-lingual Adaptation of a CTC-based multilingual Acoustic Model

Hervé Bourlard, Philip Neil Garner, Sibo Tong

Multilingual models for Automatic Speech Recognition (ASR) are attractive as they have been shown to benefit from more training data, and better lend themselves to adaptation to under-resourced languages. However, initialisation from monolingual context-de ...

ELSEVIER SCIENCE BV2018

High-dimensional peaks-over-threshold inference

Anthony Christopher Davison, Raphaël Gérard Théodore Michel Marie de Deloÿe et Fourcade de Fondeville

Max-stable processes are increasingly widely used for modelling complex extreme events, but existing fitting methods are computationally demanding, limiting applications to a few dozen variables. r-Pareto processes are mathematically simpler and have the p ...

2018

Graph Chatbot

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.