Most successes so far in deep learning have been based on supervised learning. The holy grail, however, is unsupervised learning.

“We need to go deeper.” Leonardo DiCaprio, in the film Inception (2010).

Once, artificial intelligence (AI) was science fiction. Today, it is part of our everyday lives. In the future, will computers begin to think for themselves?

Field Examples

A Complementary Monitoring Method?

Ekofisk, Norwegian North Sea, was one of the first fields where time-lapse seismic proved an excellent tool for monitoring and mapping geomechanical changes in a producing field

How can a tiny mineral called magnetite help to unravel hydrocarbon seepage and subsurface petrology? It’s not MAGIC – it’s MAGNETICS… in integrated workflows!

Simultaneous Source Separation

Decoding Nature’s Disorder

Quantum Computers

PGS launched the ‘ghost-free’ GeoStreamer GeoSource solution in 2011: a time- and depth-distributed source using sub-sources operated at specific depths and time delays, attacking the source-ghost effect in seismic data.

Can an unexplored sedimentary basin be unravelled by gravity? How can gravity in cross-disciplinary workflows estimate the base of salt domes and gas saturation in shallow sedimentary traps ? Read on!

Many TOP500 supercomputers today use both CPUs and GPUs to give the best of both worlds: GPU processing to perform mathematically intensive computations on very large data sets, and CPUs to run the operating system and perform traditional serial tasks. CPU-GPU collaboration is necessary to achieve high-performance computing.

As geophysicists, we learnt and believed that seismic deghosting through ghost notches in pressure-only recordings was impossible when the sensors were deep. CGG changed that way of thinking. Instead of fighting the ghost in acquisition, they made a different set of rules to fully exploit ‘notch diversity’ in data processing and imaging.

Energy companies need ever larger computers. This article is the second of three articles giving an introduction to supercomputers . Here, we look at the design of parallel software.