De Zarza i Cubero

I. de Zarza i Cubero

I've had several research appointments, scilicet at the Department of Computer Science and Engineering at CUHK and at Carnegie Mellon. Forbye at the EE and CS Departments at City University of Hong Kong.

I was a doctoral student in the Laboratory of Computer Vision at ETH Zürich. Previously I completed my master (with distinction) in Electrical Engineering at City University of Hong Kong, where I did an exchange at the School of Computer Science at Carnegie Mellon. I developed my master thesis at the ML Dept. and as part of the MS program, I also interned at the Robotics.

I hold a degree in Mathematics from Universitat Autònoma de Barcelona and Universitat de Barcelona. I had perfect scores in the baccalaureate and top nationwide grades in the university entrance examinations.

e-mail / cv / scholar.google / twitter

Research

My research interests span a wide range of topics related to computer vision and machine learning. That said, I enjoy both theoretical work and practical applications. Currently, in a quest to solve abstract reasoning. The ultimate goal is to build thinking conscious machines that evolve and learn with time as we humans do. Representative publications are highlighted.

	Doctor of Crosswise: Reducing Over-parametrization in Neural Networks. Curtó, Zarza, Kitani, King and Lyu. code Dr. of Crosswise proposes a new architecture to reduce over-parametrization in Neural Networks. It introduces an operand for rapid computation in the framework of Deep Learning that leverages learned weights.
	High-resolution Deep Convolutional Generative Adversarial Networks. Curtó, Zarza, Torre, King and Lyu. dataset / supplement / video In order to boost network convergence of DCGAN and achieve good-looking high-resolution results we propose a new layered network, HDCGAN, that incorporates current state-of-the-art techniques for this effect.
	Segmentation of Objects by Hashing. Curtó, Zarza, Smola and Gool. We propose a novel approach to address the problem of Simultaneous Detection and Segmentation. We use an efficient and accurate procedure that exploits the feature information of the hierarchy using Locality Sensitive Hashing.
	McKernel: A Library for Approximate Kernel Expansions in Log-linear Time. Curtó, Zarza, Yang, Smola, Torre, Ngo and Gool. code / slides / coverage McKernel introduces a framework to use kernel approximates in the mini-batch setting with Stochastic Gradient Descent (SGD) as an alternative to Deep Learning.