Hola,
¿eres nuevo aquí?

Registrate y conecta tu empresa con financiación pública, partners y proyectos.

BeyondBlackbox

Financiado

Data Driven Methods for Modelling and Optimizing the Empirical Performance of De...

Data Driven Methods for Modelling and Optimizing the Empirical Performance of Deep Neural Networks Deep neural networks (DNNs) have led to dramatic improvements of the state-of-the-art for many important classification problems, such as object recognition from images or speech recognition from audio data. However, DNNs are also... ver más

31/12/2021

ALU-FR

1M€

Presupuesto del proyecto: 1M€

Líder del proyecto

ALBERTLUDWIGSUNIVERSITAET FREIBURG No se ha especificado una descripción o un objeto social para esta compañía.

TRL 4-5

Ver 2 Participantes

Financiación concedida El organismo H2020 notifico la concesión del proyecto el día 2021-12-31

Línea de financiación objetivo El proyecto se financió a través de la siguiente ayuda:

ERC-2016-STG: ERC Starting Grant

I+D

Cerrada hace 9 años

0% 100% 100%

2 Participantes

ALU-FR

1.50M€ | Lider

MAXION WHEELS ESPAÑA, U.

335.94K€ | Participante

Proyectos interesantes

BES-2016-077626 MACRO APRENDIZAJE Y DISEÑO CON OPTIMIZACION BAYESIANA POR EN... 93K€ Cerrado

FJC2019-039023-I Optimization models in machine learning 50K€ Cerrado

DeepLearning 2.0 DeepLearning 2.0: Meta-Learning Qualitatively New Components 2M€ Cerrado

Rand4TrustPool Random sampling for trustworthy pooling layers in graph neur... 227K€ Cerrado

DYNASTY Dynamics-Aware Theory of Deep Learning 1M€ Cerrado

A-B-C-Deep Algorithmic Bias Control in Deep learning 2M€ Cerrado

Últimas noticias

29-11-2024: IDAE En las últimas 48 horas el Organismo IDAE ha otorgado 4 concesiones

29-11-2024: ECE En las últimas 48 horas el Organismo ECE ha otorgado 2 concesiones

29-11-2024: CMVAMADRID En las últimas 48 horas el Organismo CMVAMADRID ha otorgado 37 concesiones

Duración del proyecto: 61 meses Fecha Inicio: 2016-11-25
Fecha Fin: 2021-12-31

Líder del proyecto

ALBERTLUDWIGSUNIVERSITAET FREIBURG

TRL 4-5

Presupuesto del proyecto 1M€

Descripción del proyecto Deep neural networks (DNNs) have led to dramatic improvements of the state-of-the-art for many important classification problems, such as object recognition from images or speech recognition from audio data. However, DNNs are also notoriously dependent on the tuning of their hyperparameters. Since their manual tuning is time-consuming and requires expert knowledge, recent years have seen the rise of Bayesian optimization methods for automating this task. While these methods have had substantial successes, their treatment of DNN performance as a black box poses fundamental limitations, allowing manual tuning to be more effective for large and computationally expensive data sets: humans can (1) exploit prior knowledge and extrapolate performance from data subsets, (2) monitor the DNN's internal weight optimization by stochastic gradient descent over time, and (3) reactively change hyperparameters at runtime. We therefore propose to model DNN performance beyond a blackbox level and to use these models to develop for the first time: 1. Next-generation Bayesian optimization methods that exploit data-driven priors to optimize performance orders of magnitude faster than currently possible; 2. Graybox Bayesian optimization methods that have access to -- and exploit -- performance and state information of algorithm runs over time; and 3. Hyperparameter control strategies that learn across different datasets to adapt hyperparameters reactively to the characteristics of any given situation. DNNs play into our project in two ways. First, in all our methods we will use (Bayesian) DNNs to model and exploit the large amounts of performance data we will collect on various datasets. Second, our application goal is to optimize and control DNN hyperparameters far better than human experts and to obtain: 4. Computationally inexpensive auto-tuned deep neural networks, even for large datasets, enabling the widespread use of deep learning by non-experts.

Conecta tu I+D

Entra hoy

Financiación

Empresas

CTIs/Universidades

Proyectos

Investigadores