# Agustinus Kristiadi

## About

Agustinus Kristiadi is a postdoctoral fellow at the Vector Institute, working primarily with Alán Aspuru-Guzik and Pascal Poupart. He obtained his PhD from the University of Tuebingen in Germany, advised by Philipp Hennig and Matthias Hein. His research interests are in probabilistic inference with large-scale neural networks, decision-making under uncertainty, and their applications in broader scientific fields such as chemistry. His work has been recognized in the form of best PhD thesis award and multiple spotlight papers from flagship machine learning conferences. His contributions to the scientific society includementoring underrepresented students in Canada under the IBET PhD Project and actively contributing to the open-source community.

## Interests

## Experience

# Vector Institute

## Postdoctoral Fellow, Feb 2023 -

# University of Tuebingen

## PhD, Computer Science, Jun 2019 - Jan 2023

# University of Bonn

## MSc, Computer Science, Apr 2017 - Apr 2019

# GDP Venture

## Software Engineer, Apr 2013 - Dec 2015

# Universitas Atma Jaya Yogyakarta

## BEng, Software Engineering, Aug 2009 - Jan 2013

## Awards

# Best PhD thesis

## German Research Foundation's Theoretical Foundations of Deep Learning program, 2023

# Spotlight paper (top 4%)

## NeurIPS 2023

# Spotlight paper (top 3%)

## NeurIPS 2021

# Long-talk paper (top 6%)

## UAI 2021

# Best reviewer (top 10%)

## ICML 2021

## Skills

### Languages

### Editor Stack

## Latest Posts

- Writing Advice for Fledging Machine Learning Researchers
Some bullet-point advice regarding paper-writing I would give to my younger self.

- The 'use' Expression in Gleam
How can we emulate the behavior of Python's `with` and Rust `?` in Gleam?

- Volume Forms and Probability Density Functions Under Change of Variables
From elementary probability theory, it is well known that a probability density function (pdf) is not invariant under ...

- The Invariance of the Hessian and Its Eigenvalues, Determinant, and Trace
In deep learning, the Hessian and its downstream quantities are observed to be not invariant under reparametrization. This ...

- Convolution of Gaussians and the Probit Integral
Gaussian distributions are very useful in Bayesian inference due to their (many!) convenient properties. In this post we ...

## Selected Works

- Being Bayesian, Even Just a Bit, Fixes Overconfidence in ReLU NetworksAgustinus Kristiadi, Matthias Hein, Philipp Hennig