Výsledky vyhledávání - "Benatan, Matt"

Report

An adaptive approach to Bayesian Optimization with switching costs

Autor: Pricopie, Stefan, Allmendinger, Richard, Lopez-Ibanez, Manuel, Fare, Clyde, Benatan, Matt, Knowles, Joshua

We investigate modifications to Bayesian Optimization for a resource-constrained setting of sequential experimental design where changes to certain design variables of the search space incur a switching cost. This models the scenario where there is a

Externí odkaz: http://arxiv.org/abs/2405.08973

Zobrazit plný text záznamu

Kniha

Enhancing Deep Learning with Bayesian Inference : Create More Powerful, Robust Deep Learning Systems with Bayesian Deep Learning in Python. [elektronicky zdroj]

Autor: Benatan, Matt

Externí odkaz: Kolekce e-knih KNAV (Registrovani uzivatele: plny text online 5 minut, dalsi pristup na vyzadani. Registered users: full text online 5 minutes, further access on request.)

Report

Model-agnostic variable importance for predictive uncertainty: an entropy-based approach

Autor: Wood, Danny, Papamarkou, Theodore, Benatan, Matt, Allmendinger, Richard

In order to trust the predictions of a machine learning algorithm, it is necessary to understand the factors that contribute to those predictions. In the case of probabilistic and uncertainty-aware models, it is necessary to understand not only the r

Externí odkaz: http://arxiv.org/abs/2310.12842

Zobrazit plný text záznamu

Report

SonOpt: Sonifying Bi-objective Population-Based Optimization Algorithms

Autor: Asonitis, Tasos, Allmendinger, Richard, Benatan, Matt, Climent, Ricardo

We propose SonOpt, the first (open source) data sonification application for monitoring the progress of bi-objective population-based optimization algorithms during search, to facilitate algorithm understanding. SonOpt provides insights into converge

Externí odkaz: http://arxiv.org/abs/2202.12187

Zobrazit plný text záznamu

Report

Fully Bayesian Recurrent Neural Networks for Safe Reinforcement Learning

Autor: Benatan, Matt, Pyzer-Knapp, Edward O.

Reinforcement Learning (RL) has demonstrated state-of-the-art results in a number of autonomous system applications, however many of the underlying algorithms rely on black-box predictions. This results in poor explainability of the behaviour of thes

Externí odkaz: http://arxiv.org/abs/1911.03308

Zobrazit plný text záznamu