Borja de Balle Pigem

News: I am proud to announce that I will be joining the Department of Mathematics and Statistics and the Data Science Institute at Lancaster University as a lecturer in October.

News: I am Workshops co-Chair for NIPS 2015.

I am currently a post-doctoral fellow in the Reasoning and Learning Laboratory at McGill University under the supervision of Prakash Panangaden, Joelle Pineau, and Doina Precup. I recently co-organized workshops on Spectral Learning and the Method of Moments at ICML 2014, NIPS 2013 and ICML 2013. I was also Area Chair for NIPS 2014 and member of the Program Commitees for IJCAI 2015 (Machine Learning Track) and ICGI 2014.

I obtained my PhD in 2013 from UPC after working in the LARCA research group under the supervision of Jorge Castro and Ricard Gavaldà. During my PhD I spent several months visiting Mehryar Mohri at the Courant Institute (NYU).

Before this, I got a MsC in Mathematics from the FME and wrote a master's thesis on Combinatorial Group Theory advised by Enric Ventura. During my undergrad years I studied Telecommunication Engineering and Mathematics at CFIS and did a final year project on Linear Feedback Shift Registers directed by Josep M. Fuertes and Enric Ventura, for which I won an award from the SCT.


McGill University
McConnell Engineering Building, Room 104N
3480 University st., Montreal, QC, H3A2A7, Canada


bballe /at/ cs /dot/ mcgill /dot/ ca

My research interests revolve around all aspects of Machine Learning: theory, algorithms, and applications.

A lot of my recent work involves scalable algorithms based on spectral or greedy principles for learning latent-variable models inspired by Language Theory and Dynamical Systems.

I like to design algorithms that can solve real-world problems arising in Natural Language Processing and Reinforcement Learning.

Papers available here may be subject to copyright and are intended for personal, non-commercial use only.

B. Balle, J. Castro, and R. Gavaldà
Learning Probabilistic Automata: A Study In State Distinguishability
Theoretical Computer Science, 473:46-60, 2013  (DOI)

B. Balle and M. Mohri
Spectral Learning of General Weighted Automata via Constrained Matrix Completion
Neural Information Processing Systems (NIPS), 2012
(Honorable Mention for the Outstanding Student Paper Award)

B. Balle, J. Castro, and R. Gavaldà
Bootstrapping and Learning PDFA in Data Streams
International Colloquium on Grammatical Inference (ICGI), 2012
(Best Student Paper Award)

B. Balle, A. Quattoni, and X. Carreras
Local Loss Optimization in Operator Models: A New Insight into Spectral Learning
International Conference on Machine Learning (ICML), 2012

F. M. Luque, A. Quattoni, B. Balle, and X. Carreras
Spectral Learning for Non-Deterministic Dependency Parsing
Conference of the European Chapter of the Association for Computational Linguistics (EACL), 2012
(Best Paper Award)

B. Balle, A. Quattoni, and X. Carreras
A Spectral Learning Algorithm for Finite State Transducers
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2011

B. Balle, J. Castro, and R. Gavaldà
A Lower Bound for Learning Distributions Generated by Probabilistic Automata
International Conference on Algorithmic Learning Theory (ALT), 2010

B. Balle
Implementing Kearns-Vazirani Algorithm for Learning DFA Only with Membership Queries
Zulu Workshop, 2010
(The algorithm described in this paper finished in 2nd place in the Zulu Competition)

B. Balle, J. Castro, and R. Gavaldà
Learning PDFA with Asynchronous Transitions
International Colloquium on Grammatical Inference (ICGI), 2010

B. Balle, E. Ventura, and J.M. Fuertes
An Algorithm to Design Prescribed Length Codes for Single-Tracked Shaft Encoders
IEEE International Conference on Mechatronics (ICM), 2009

J.M. Fuertes, B. Balle, and E. Ventura
Absolute-Type Shaft Encoding Using LFSR Sequences With a Prescribed Length
IEEE Transactions on Instrumentation and Measurement, Vol. 57, No. 5, 2008

Tutorial on Spectral Learning Techniques for Weighted Automata, Transducers, and Grammars   —   Empirical Methods in Natural Language Processing (EMNLP), October 2014

Spectral Methods for Learning Finite State Machines   —   McGill University, September 2012

Recuperació de la Informació   —   FIB, Primavera 2011

Teoria de la Computació   —   FIB, Maig 2011

Entropi Fylms   —   Ràdio 90

Aiki O Kami   —   Bond Street Dojo   —   Ritzu Zen Garrotxa   —   McGill Aikido