….
SUPERCOMPUTING and ARTIFICIAL INTELLIGENCE Research, Teaching and Scientific Dissemination
Professor at UPC Barcelona Tech
Research Manager at Barcelona Supercomputing Center
FEATURED INFORMATION
NUEVO LIBRO EN CASTELLANO
SOBRE aprendizaje por refuerzo
INTRODUCCIÓN AL APRENDIZAJE POR REFUERZO Teoría y práctica en python
Deep Reinforcement Learning Explained
A series that gradually and with a practical approach will be introducing the reader in this exciting technology of Deep Reinforcement Learning.
Transfer Learning: La democratització dels Transformers
La llibreria Hugging Face posa els Transformers a l’abast de tothom (versió anglesa d’aquesta publicació a Medium) Ja hem vist que les xarxes neuronals Transformers son actualment la tecnologia dominant en l’àmbit del processat de llenguatge natural, podent-se usar en tasques com classificació, extracció, generació, etc. Per exemple, permet classificar un text en funció de
13. Accelerate the learning with Parallel training on multiple GPUs
Hands-on description Deep learning codes are highly parallel, which makes them not only conducive to taking advantage of GPU acceleration but also scalable to multiple GPUs and multiple nodes. In today’s hands-on, we will present how we could paralyze a single Deep Neural Network training over many GPUs in one node of CTE-POWER using tf.distributed.MirroredStrategy()
12. Parallel and Distributed Deep Learning Frameworks
Hands-on exercise description In today’s hands-on exercise we will review the basic concepts and an image classification problem (based on the CIFAR10 dataset) using a neural network model as a classifier (the neural network ResNet50V2 with 25,613,800 parameters). We will introduce and analyze all the required steps to train this classifier sequentially using one GPU. This
Transformers: La nova joia del “Deep Learning”
(versió anglesa d’aquesta publicació a medium) La major part del nostre temps ens comuniquem en llenguatge natural, és a dir, en text no estructurat, ja sigui escrit o parlat. Per aquest motiu, des de fa molts anys s’està treballant per aconseguir que les màquines comprenguin aquests textos no estructurats i n’extreguin la informació rellevant. És
11. Using Supercomputers for DL training
Last Monday, all the attendees (including the teacher) agreed that today’s class could be online in our meet room, utilising the hands-on exercise described below. The experience will be as a proof of concept class to know if we can follow the remaining classes (of this part 2 of the course) without requiring our presence
Without the Rise of Supercomputers There Will Be No Progress in Artificial Intelligence
We need solutions that overlap these two research worlds Although artificial intelligence (AI) research began in the 1950s, and the fundamental algorithms of neural networks were already known in the last century, this discipline was left dormant for many years due to the lack of results that could excite both academia and the industry. Artificial Intelligence