TF2, Physics in GCNs and Aftershocks – Friday Faves

In this week’s Friday Faves, we have Tensorflow 2 dropping and a beautiful bonus, next level physics-based ML, and a problem with a Harvard deep learning paper. Tensorflow 2 for Researchers Tensorflow 2.0 dropped this week and it has Eager Execution (read “normal behaviour”) per default and the Keras API per default. If you’re familiar […]

GIS, Fossils and Loads of Data – Friday Faves

This week, we have loads of geoscience-y favourites. Between great Jupyter widgets, augmented reality and a huge new dataset to play with, in the Friday Faves. Geoscience and Python Martin Renou gave a fantastic talk about Jupyter Voila at EuroScipy. He tweeted out their newest developments in iPyLeaflet, which brings GIS capabilities to Jupyter. Once […]

A Week in Tech Heaven – Googling in San Francisco [1/3]

So am I understanding this correct: You are here for a $1600 conference. And it’s not for work. This is your free time? Yes. I recently travelled to San Francisco to attend the Google Next conference. Clearly neither a scientific conference nor related to geophysics. And also clearly, the immigration officer was not particularly on […]

Research Talk — Deep Learning for 4D Pressure Saturation Inversion [Youtube]

I presented in Amsterdam during the Practical Reservoir Monitoring Workshop in Amsterdam. This is the accompanying video. Abstract In this work, we present a deep neural network inversion on map-based 4D seismic data for pressure and saturation. We present a novel neural network architecture that trains on synthetic data and provides insights into observed field […]

Keynote Bonanza and No Coffee – The EAGE / PESGB ML Workshop

Last month EAGE and PESGB organized the first machine learning workshop in geoscience in Europe. Clearly, I had every intention of going. And obviously, I met many of my favourite co-conspirators there, when I did. The workshop was divided between a day of keynotes and a day of technical talks. The keynotes accompanied the PETEX […]

Cheaper Deep Learning by Transfer Learning Cats to Seismic [SEG Conference 2018]

This years SEG I got the opportunity to present some of my work on transfer learning. In automatic seismic interpretation, progress is preceived as incremental already, although the field has only been established fairly recently. It was shown with new deep neural networks, usually convolutional neural networks, that reproducing human seismic interpretation is possible. This […]

Culture Neutral - CC-BY Karl-Ludwig Poggemann

Data Is Not Neutral

  I like data driven processes. You’ve seen me drum on about CRS and Full-Waveform Inversion. Sometimes we like to pretend that data is pure. Well , it’s noisy and full of kinks but if we process it just right, we extract the truth. Turns out that is a little bit wrong. The data collection […]

Dipping your Toes – Machine Learning for Geoscientists

My fellow students know this and I hope recruiters will never read this: I was never good at math in university. It was only later when it came to the application in actual geophysical problems that tensors, linear algebra, and differential equations clicked. Personally, I don’t recommend this, as it makes life unnecessarily hard. Machine […]