Ghifari Adam Faza

Ghifari Adam Faza

PhD Student at KU Leuven

Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Posts

Simple robust optimization tutorial

7 minute read

Published:

Let’s say that you are working in a local taxi company where they want to develop their in-house navigation software. This navigation software is responsible for showing the taxi drivers the optimum directions to the destination or pick-up points. Naturally, arriving quickly is crucial—after all, as the saying goes, time is money!

Tutorial penyelesaian Markov Decision Process dengan pemrograman linier 🇮🇩

17 minute read

Published:

Pada postingan kali ini, kita membahas implementasi dari Markov decision process (MDP) sebagai alat untuk menyelesaikan permasalahan proses pengambilan keputusan dari sistem dinamik dengan memanfaatkan metode pemrograman linier. Pertama, kita akan membahas definisi dari MDP secara singkat. Lalu, kita akan melihat sebuah contoh kasus dari MDP untuk menentukan kebijakan optimal (optimal policy) dari perawatan mesin industri. Berdasarkan contoh kasus ini, kita akan membahas cara untuk menformulasikan representasi MDP yang sesuai. Terakhir, kita akan menyelesaikan representasi MDP tersebut dengan menggunakan teknik pemrograman linier untuk memperoleh keputusan optimal dari proses tersebut.

Tutorial on solving Markov Decision Process with linear programming

18 minute read

Published:

In this post, we discuss the hands-on implementation of the Markov decision process (MDP) as a tool to solve the decision-making process of a dynamic system by leveraging the linear programming method. First, we will briefly discuss the definition of MDP. Then, we will consider a use case of MDP to determine the optimal policy for industrial machine maintenance. Based on this use case, we will discuss how to formulate a suitable MDP representation. Finally, we will solve the MDP representation through linear programming techniques to obtain the optimal policy for the decision-making process.

Prediksi Bidang Vektor (Vector Field) Sederhana dengan Proper Orthogonal Decomposition dan Polynomial Chaos Expansion (POD-PCE) 🇮🇩

16 minute read

Published:

Dalam artikel ini, kita mengeksplorasi penggunaan proper orthogonal decomposition (POD) dan polynomial chaos expansion (PCE) sebagai surrogate model untuk memprediksi perilaku bidang vektor pada rentang waktu sembarang. Dibandingkan dengan metode surrogate model standar lainnya seperti regresi PCE standar atau regresi Gaussian process (GP) standar, penambahan komponen POD memiliki kelebihan untuk menganalisis data dalam bidang vektor dengan mentransformasi bidang vektor seperti aliran fluida pada video dibawah menjadi representasi yang lebih ringkas.

Simple Vector Field Prediction using Proper Orthogonal Decomposition and Polynomial Chaos Expansion (POD-PCE)

19 minute read

Published:

In this article, we explore the utilization of proper orthogonal decomposition (POD) and polynomial chaos expansion (PCE) as a surrogate model to forecast flow field behavior at unseen time steps. When compared to other standard surrogate models like standard PCE regression or standard Gaussian process (GP) regression, the integration of POD presents advantages in the analysis of vector-field data, by converting the complete solution of a vector field into a collection of simpler representations that are more manageable.

portfolio

Epistemic AI - EU Horizon 2020 Project


Although artificial intelligence (AI) has improved remarkably over the last years, its inability to deal with fundamental uncertainty severely limits its application. This proposal re-imagines AI with a proper treatment of the uncertainty stemming from our forcibly partial knowledge of the world. As currently practiced, AI cannot confidently make predictions robust enough to stand the test of data generated by processes different (even by tiny details, as shown by ‘adversarial’ results able to fool deep neural networks) from those studied at training time. While recognising this issue under different names (e.g. ‘overfitting’), traditional machine learning seems unable to address it in nonincremental ways. As a result, AI systems suffer from brittle behaviour, and find difficult to operate in new situations, e.g. adapting to driving in heavy rain or to other road users’ different styles of driving, e.g. deriving from cultural traits. Epistemic AI’s overall objective is to create a new paradigm for a next-generation artificial intelligence providing worst-case guarantees on its predictions thanks to a proper modelling of real-world uncertainties.More information…

KADAL


Kriging for Analysis, Design optimization, And expLoration (KADAL) is a Python package that integrates a wide range of Bayesian Optimization tools. These include various surrogate modeling methods, sampling techniques, optimization methods, and tools for uncertainty analysis such as uncertainty quantification, sensitivity analysis, and reliability analysis. KADAL provides a comprehensive solution for analyzing, optimizing, and exploring complex systems with uncertainty.

mySVR

A standard support vector regression with multi-kernel capability, suitable for educational purposes.

Sobol Sampling

Python package for generating Sobol sequences, able to generate up to 21201 dimension.

publications

Interval Reduced Order Surrogate Modelling Framework for Uncertainty Quantification

Published in AIAA SciTech 2024 , 2024

This paper introduces a non-intrusive framework for epistemic surrogate modeling, leveraging interval proper orthogonal decomposition (interval POD) and interval polynomial chaos expansion (interval PCE) to handle interval observations, addressing a major limitation in existing frameworks. By integrating POD for interval data with PCE for interval observations, the framework allows for the consideration of non-scalar data, such as intervals, providing a more comprehensive approach to physical system modeling that captures additional information.

Download here

talks

teaching