Juan Raúl Padrón Griffe

About me
I am a Marie Sklodowska-Curie Fellow of the EU Project PRIME and a PhD candidate at the Graphics and Imaging Lab. My PhD thesis under the supervision of Prof. Adolfo Muñoz and Prof. Adrian Jarabo focuses on physically-based rendering and appearance modeling of multi-scale materials, such as biological tissues (skin, scales and feathers) and intricate human-made objects (cosmetics). Previously, I earned my Bachelor of Sciene degree in Computer Science at the Universidad Central de Venezuela, where I specialized in computer graphics and imaging processing. My undergraduate thesis explored the generation and visualization of procedural terrains. Later, I received my Master of Science degree in Informatics at the Technical University of Munich, concentrating on computer graphics, computer vision and machine learning. During my Master studies, I conducted research on 3D Scanning and Neural Rendering for object and face relighting advised by Dr. Justus Thies. Beyond my academic experience, I have two years of software development experience in backend technologies (.NET, Service Stack, Java, Spring).

Looking for Opportunities
I recently submitted my Ph.D. dissertation under the title “Modeling and Rendering of Multiscale Materials”. I am currently seeking both postdoctoral and industry opportunities where I can apply my expertise in computer graphics, computer vision and artificial intelligence for the digital acquisition, representation and understanding of the visual world. My combined expertise in computer graphics, computer vision, machine learning, and software engineering allows me to tackle complex technical challenges from both a research and implementation perspective, If you're interested in collaboration or have an opportunity that aligns with my expertise, please feel free to reach out!

Projects

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

Presentation

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.

Twitter
# Protein Folding # Reinforcement Learning