TensorFlow counterattacks to Caffe2

May 4th, 2017|

Few days ago appeared Caffe2 as a new player on the scene of Deep Learning frameworks with impressive performance scaling results on DGX-1. Tonight, Tensorflow shown the same numbers and the GitHub code to prove it yourself: My last post finished with this Q&A: "Will Caffe2 steal the supremacy that TensorFlow has right now? I do not have a crystal ball. My opinion is that It seems not easy, but it is the only one that could be a competitor for Google especially in the production arena in companies. We will see it soon, because Facebook has to hurry, the window of time to get it will be very small." Let me add: The window is about to close if Facebook does not work fast and hard!   UPDATE 9-May (thanks to Maurici Yagües): https://twitter.com/soumithchintala/status/861944937100722176  :-)       [...]

Caffe2: A new player on the scene of Deep Learning frameworks

April 25th, 2017|

Few days ago Facebook announced Caffe2, a new open-source, cross-platform framework for deep learning. They say that Caffe2 is the successor to Caffe ( really?), the deep learning framework developed by Berkeley AI Research and community contributors. Caffe2’s GitHub page describes it as “an experimental refactoring of Caffe that allows a more flexible way to organize computation.” As my readers know, when appeared TensorFlow I decided to pay attention to it because it could change the scene of DL/AI frameworks. Now, we are in the same situation, Caffe2 could change the current scene that Francesc Sastre, one of my master students,  build for his master thesis:  "Frameworks popularity evolution in GitHub" No questions, right? Facebook launched Caffe2, an open-source deep learning framework made with expression, speed, and modularity in mind. It address the bottlenecks observed [...]

Artificial Intelligence & Barça

March 22nd, 2017|

Ayer participé en la jornada de formación interna “Ciencia y Tecnología aplicada", dirigida a directivos del área de innovación y cuerpo técnico del FC Barcelona, hablando de Inteligencia Artificial. En la sesión también participó Josep Lladós, director del Centre de Visió per Computador (CVC) y José Ignacio Latorre, catedrático de Física Cuántica de la Universitat de Barcelona.  Fantásticas presentaciones las de Josep y José Ignacio. Viéndolos, uno se da cuenta de lo mucho que siempre queda por aprender. Espero que la gente del Barça sacara valor de los contenidos de la jornada. Gracias por invitar al BSC-CNS. Tal como me comprometí adjunto (un poco tarde ;-) ) las transparencias de la presentación.  También adjunto unas fotos que me tomo Anna Escoda del BSC-CNS. Gracias Anna. Artificial Intelligence overview from Barcelona Tech UPC [...]

Deep Learning at UPC with Oriol Vinyals & Joan Bruna

May 23rd, 2016|

Today we had a great event, the first seminar on Deep Learning organised by UPC, TelecomBCN, FIB, CFIS and BSC. Thanks Joan and Oriol, It has been impressive!  Enclosed you will find some pictures of his visit! Here you can download the slides of Joan Bruna and here the video of the talk. As soon as I receive the presentation of Oriol I will post the links here too. Huge audience! (photo Sebastià Xambó) (photo Joan Bruna) (photo GabrielOliveira) Two great speakers: (photo Sebastià Xambó) (photos Joan Capdevila) (photos Jordi Torres) Presenters: (photos Joan Capdevila) Meetings with UPC students:(photo Xavier Giro-i-Nieto) Marenostrum visit: (photo Jordi Torres) With (his) former Dean of the Faculty of Mathematics and Statistics (photo Sebastià Xambó)

Seminar on Deep Learning at UPC with Oriol Vinyals and Joan Bruna

May 14th, 2016|

We (UPC, TelecomBCN, FIB, CFIS and BSC) are proud to invite you to our first seminar on Deep Learning organized at UPC, next May 23th at 11:00, with the talks of two outstanding speakers: Joan Bruna (UC Berkeley) and Oriol Vinyals (Google DeepMind). Do not miss this great event happening in Barcelona.  And if you are interested in Deep Learning, we are also preparing  a Deep Learning for Computer Vision Summer Seminar  in Barcelona next 4-8 July 2016 ( more info at  http://TelecomBCN.DeepLearning.Barcelona). UPDATE: Photographs of the event here When: Monday  May 23, 11:00-13:00 Place: Aula Màster -  A3 building - UPC Campus Nord - Barcelona Talks: 11:00: Convolutional Neural Networks against the Curse of Dimensionality by Joan Bruna (UC Berkeley).  Convolutional Neural Networks are a powerful class of non-linear representations that have shown through numerous [...]

Free Deep Learning Books

April 18th, 2016|

There are not many books on Deep Learning at the moment because it is such a young area of study. However there are new freely online version available of 3 books that cover beginner, intermediate and advanced levels:   Advanced level: "Deep Learning" written by top deep learning scientists Ian Goodfellow, Yoshua Bengio and Aaron Courville. The book includes coverage of all of the main algorithms in the field and can be used as a reference book by those who have experience in the field. Intermediate level: "Neural Networks and Deep Learning" written by Michael Nielsen. The book is a good place to start for those who already have experience in Machine Learning and want to delve into Neural Networks. Beginner level: "FIRST CONTACT WITH TENSORFLOW, get started with Deep Learning programming" written by Jordi Torres. The book is [...]

Aprender TensorFlow en Salamanca

March 31st, 2016|

Hace unos días unos alumnos de la Facultad de Ciencias de la USAL pertenecientes al  capítulo de la ACM se pusieron en contacto conmigo a propósito del libro TensorFlow y en especial por una de sus ilustraciones.  Uno de sus intereses es el campo de la inteligencia artificial y las redes neuronales artificiales, así que desde que fundaron la asociación han venido organizando anualmente workshops con esta temática [1][2]. Ahora organizan un seminario introductorio al aprendizaje automático con redes neuronales y TensorFlow. Me han mostrado su github que han preparado para el curso y sin duda va a ser impresionante, tratando temas  como las LSTM que no contiene mi libro. Les pregunté que debería contar a mis alumnos para animarlos a asistir a este seminario en Salamanca. Aquí va: "La verdad es que estudiar en Salamanca es toda una [...]

HOY, UN HUMANO AÚN HA PODIDO SUPERAR AL COGNITIVE COMPUTING, ¿HASTA CUÁNDO?

March 13th, 2016|

Hoy, el surcoreano Lee Sedol, campeón mundial del juego de mesa GO, ha logrado su primera victoria contra la aplicación de “Cognitive Computing” AlphaGo  de Google. Ha sido en la cuarta de las cinco partidas de las que consiste el torneo que se realiza en Seúl (Corea del Sur). Lee Sedol, en realidad, ya perdió el torneo después de que AlphaGo ganara las tres primeras partidas, pero el hecho que hoy Lee haya ganado al sistema es noticia. Si les apeteciera, el próximo martes pueden seguir en directo la final en YouTube. Según un tweet de Demis Hassabis,  CEO y cofundador de DeepMind (compañía de inteligencia artificial inglesa adquirida por Google en 2014), el sistema AlphaGo se confundió y no se dio cuenta que había cometido un error hasta diez movimientos después. Pero [...]

Google launched TensorFlow Serving

February 18th, 2016|

Google launched TensorFlow Serving, that helps developers to take their TensorFlow machine learning models (and, even so, can be extended to serve other types of models) into production.  TensorFlow Serving is an open source serving system (written in C++) now available on GitHub under the Apache 2.0 license. What is the difference between TensorFlow and TensorFlow Serving?  While in TensorFlow is easier for the developers to build machine learning algorithms and train them for certain types of data inputs, TensorFlow Serving specializes in making these models usable in production environments.  The idea is that developers train their models using TensorFlow and then they use TensorFlow Serving’s APIs to react to input from a client. This allows developers to experiment with different models in a large scale that change over time based on real-world data, and maintain a [...]

INTERNATIONAL PHD IN DEEP LEARNING & HPC FOR COMPUTER VISION

February 7th, 2016|

The Research Group Autonomic Systems and eBusiness Platforms at Barcelona Supercomputing Center (BSC), invites outstanding candidates to apply for a full-time PhD position under "Becas de posgrado de La Caixa"  grants (link) or "International PhD Program Fellowship BSC La Caixa" grants (link). The PhD work will focus on Deep Learning and High Performance Computing for computer vision. PhD project description Deep learning has changed the scape of computer vision and artificial intelligence. Current research in these areas has shown the possibility of creating tools that are able to recognize thousands of objects, or automatically describe the content in an image. For example, Google Brain approach to automatic captioning is able to reliably describe many complex images using this basic queues. However, human understanding of images is far more elaborated. Humans do not [...]

Acto presentación primer libro sobre TensorFlow

February 2nd, 2016|

Primeras fotos del acto de presentación del primer libro que se ha escrito sobre el paquete TensorFlow, a la venta en Amazon.es y próximamente disponible en esta web (cuando el autor encuentre un momento!). Gracias a todos por vuestra asistencia, espero que lo pasarais tan bien como yo. De verdad que agradezco muchísimo vuestra compañía, ¡Gracias!   LINK AL LIBRO CALÇOTADA SEASON AND THE NEW DEEP LEARNING BOOK Saint George's Day and the Book: