Rafael Cabañas: Deep Probabilistic Modelling made easy
28 May 2019
Manno, Galleria 1, 2nd floor, room G1-201 @12:00
InferPy is an open-source library for deep probabilistic modeling written in Python and running on top of Edward 2 and Tensorflow. Other existing probabilistic programming languages possess the drawback that they are difficult to use, especially when defining deep neural networks and probability distributions over multidimensional tensors. This means that their final goal of broadening the number of people able to code a machine learning application may not be fulfilled. InferPy tries to address these issues by defining a user-friendly API which trades-off model complexity with ease of use. In particular, this library allows users to: prototype hierarchical probabilistic models with a simple and user-friendly API inspired by Keras; define probabilistic models with complex constructs containing deep neural networks; create computationally efficient batched models without having to deal with complex tensor operations; and run seamlessly on CPUs and GPUs by relying on Tensorflow.

The speaker

Rafael Cabañas is  a new postdoc researcher at IDSIA. In general, his research is framed in the fields of data analysis and machine learning, and more specifically related to uncertainty treatment with probabilistic graphical models (PGMs).
He did his Ph.D. in the Department of Computer Science and AI of the University of Granada (Spain). During this time, he worked in the study of new algorithms and data structures for the inference of influence diagrams, a kind of PGM for decision making. Then, he was involved in the European project AMIDST as a researcher hired by the Aalborg University (Denmark), where he worked in the development of a Java open-source library for the analysis of streaming data with PGMs.  This software is being used by two private sector companies, namely the Spanish regional bank BCC and Danish IT company Hugin Expert.
In the past few months, as a postdoc researcher at the University of Almería (Spain), he has been developing a Python library of a probabilistic programming language over GPUs, namely InferPy. Its main purpose is to provide an easy-to-use syntaxis for the definition of probabilistic programs with neural networks.


Pizza and drinks will be offered at the end of the talk. If you plan to attend, please register in a timely fashion at the following link so that we will have no shortage of food: