Juan Raúl Padrón Griffe

Epale! I am PhD candidate at the University of Zaragoza as a member of the PRIME Network with the Graphics and Imaging Lab. My PhD thesis focus on developing theory and methods for efficient rendering of volumetric heterogeneous appearances and it is being supervised by Prof. Adrian Jarabo. Previously, I obtained my Bachelor degree in Computer Science (2015) at the Central University of Venezuela. Later, I received my Master degree in Informatics (2020) at the Technical University of Munich. During my Master studies I focused mostly on the Computer Graphics and Vision subjects, where I was fortunate enough to be advised by Prof. Matthias Niessner and Dr. Justus Thies at the Visual Computing lab to research on 3D Scanning and Neural Rendering.

I am really enthusiastic about the intersection of realistic image synthesis, graphics-based vision and machine learning for the digital acquisition, representation and understanding of the visual world. In particular, I am interested on pushing the state of the art on physically-based rendering, neural rendering and inverse rendering. I also find really interesting other research areas from the Graphics and Imaging Lab such as Human Perception and Transient Imaging.


Deep Reinforcement Learning for Protein Folding

2018, Oct 02    

The protein structure determines properties and functions of the protein. Therefore, scientistic can develop drugs for this specific unique protein shape in order to cure diseases. Nowadays, Machine Learning methods arise as an alternative to the costly experimentation techniques (cryo-electron microscopy, nuclear magnetic resonance, X-ray crystallography) to help accelerate research.

For the Hands-on Deep Learning for Computer Vision practical course at the Technical University of Munich, we implement the infamous AlphaGo algorithm in C++ to address the protein folding problem with the reinforcement learning approach. The main idea of the algorithm is training a neural network to estimate the policy and the values estimates, where the policy is improved by looking ahead into the future via Monte Carlo Tree Search guided by the value network. The environment (Rosetta 2017.39) and the Monte Carlo Tree Search algorithm were implemented in C++ 17, while the neural network was written in Python (TensorFlow 1.4.1). We rely on the TensorRT library to improve significantly the performance of the neural network inference. As a result, we could speed up the implementation 18.95 times regarding to the python version. Nevertheless, the approach is still unfeasible for large proteins (e.g. L = 332).

Results overfitting a single protein:

Smooth Return Value Loss Policy Loss

Team Members: Juan Raul Padron Griffe, Matthias Humt, Felix Opolka
Instructors: Vladimir Golkov, Daniel Cremers


DeepMind implemented successfully this project idea in a system called AlphaFold. If you want to know more about cool projects in biomedicine, please contact Vladimir Golkov. Felix Opolka, which was the main author of this project, is currently working on cool publications about neural networks for graph-structured data.