Ricard Gavaldà

Reducing wasted resources to help achieve green data centers (2009)

Closed patterns are powerful representatives of frequent patterns, since they eliminate redundant information. We propose a new approach for mining closed unlabeled rooted trees adaptively from data...

Jordi Torres, David Carrera, Kevin Hogan, Ricard Gavaldà, Vicenç Beltran, Nicolás Poggi

In this paper we introduce a new approach to the consolidation strategy of a data center that allows an important reduction in the amount of active nodes required to process a heterogeneous workload...

Algorithmic Learning Theory: 20th International Conference, ALT 2009, Porto, Portugal, October 3-5, 2009. Proceedings (2009)

Gavaldà, Ricard.

Adaptive Distributed Mechanism Against Flooding Network Attacks Based on Machine Learning (2008)

Berral, Josep Ll., Poggi, Nicolas, Alonso, Javier, Gavaldà, Ricard, Torres, Jordi

Adaptive techniques based on machine learning and data mining are gaining relevance in selfmanagement and elf-defense for networks and distributed systems. In this paper, ee focus on early detection...

Towards feasible PAC-learning of probabilistic deterministic finite automata (2008)

Castro, Jorge, Gavaldà, Ricard

We present an improvement of an algorithm due to Clark and Thollard (Journal of Machine Learning Research, 2004) for PAC-learning distributions generated by Probabilistic Deterministic Finite...

Mining adaptively frequent closed unlabeled rooted trees in data streams (2008)

Machine Learning in Secondary Education? (2008)

Closed patterns are powerful representatives of frequent patterns, since they eliminate redundant information. We propose a new approach for mining closed unlabeled rooted trees adaptively from data...

Gavaldà, Ricard

Given the alarming drop in applications to Computer Science (CS) studies across Europe, any efforts at presenting CS as an interesting, challenging, and useful discipline are welcome and necessary....

Reducing wasted resources to help achieve green data centers (2008)

Torres, Jordi, Carrera, David, Hogan, Kevin, Gavaldà, Ricard, Beltrán, Vicenç, Poggi, Nicolas

In this paper we introduce a new approach to the consolidation strategy of a data center that allows an important reduction in the amount of active nodes required to process a heterogeneous workload...

computational complexity NON-AUTOMATIZABILITY OF BOUNDED-DEPTH FREGE PROOFS (2008)

Maria Luisa Bonet, Carlos Domingo, Ricard Gavaldà, Alexis Maciel, Toniann Pitassi

Abstract. In this paper, we show how to extend the argument due to Bonet, Pitassi and Raz to show that bounded-depth Frege proofs do not have feasible interpolation, assuming that factoring of Blum...

Automatic Detection and Banning of Content Stealing Bots for E-commerce (2008)

Nicolás Poggi, Josep Lluis Berral, Toni Moreno, Ricard Gavaldà, Jordi Torres

Content stealing in the web is becoming a serious concern for information and e-commerce websites. In the practices known as web fetching or web scraping [1], a stealer bot simulates a human web user...

Early Drift Detection Method (2008)

Raúl Fidalgo, Albert Bifet, Ricard Gavaldà, Rafael Morales-bueno

An emerging problem in Data Streams is the detection of concept drift. This problem is aggravated when the drift is gradual over time. In this work we define a method for detecting concept drift,...

Self-Adaptive Utility-Based Web Session Management (2008)

Poggi, Nicolas, Moreno, Toni, Berral, Josep Lluis, Gavaldà, Ricard, Torres, Jordi

In the Internet, where millions of users are a click away from your site, being able to dynamically classify the workload in real time, and predict its short term behavior, is crucial for proper...

Automatic Detection and Banning of Content Stealing Bots for E-commerce (2007)

Poggi, Nicolas, Berral, Josep Lluis, Moreno, Toni, Gavaldà, Ricard, Torres, Jordi

Content stealing in the web is becoming a serious concern for information and e-commerce websites. In the practices known as web fetching or web scraping, a stealer bot simulates a human web user to...

Computational Power of Neural Networks: A Kolmogorov Complexity Characterization (2007)

José L. Balcázar, Ricard Gavaldà, Hava T. Siegelmann

The computational power of neural networks depends on properties of the real numbers used as weights. We focus on networks restricted to compute in polynomial time, operating on boolean inputs....

Compressibility of Infinite Binary Sequences (2007)

José L. Balcázar, Ricard Gavaldà, Montserrat Hermo

It is known that infinite binary sequences of constant Kolmogorov complexity are exactly the recursive ones. Such a kind of statement no longer holds in the presence of resource bounds. Contrary to...

Learning Decision Trees Adaptively from Data Streams with Time Drift (2007)

Web Customer Modeling for Automated Session Prioritization on High Traffic Sites (2007)

We propose a new method for mining concept-drifting data streams using decision trees and adaptive windowing. We present a new algorithm based on Hulten-Spencer- Domingos’s CVFDT that overcomes...

Poggi, Nicolas, Moreno, Toni, Berral, Josep Lluis, Gavaldà, Ricard, Torres, Jordi

In the Web environment, user identification is becoming a major challenge for admission control systems on high traffic sites. When a web server is overloaded there is a significant loss of...

Policy-based autonomous bidding for overload management in eCommerce websites (2007)

Moreno, Toni, Poggi, Nicolas, Berral, Josep Lluis, Gavaldà, Ricard, Torres, Jordi

In eCommerce applications with heterogeneous traffic, which typically run on execution environments where resources are shared with other applications, being able to dynamically adapt to the...

Learning from Time-Changing Data with Adaptive Windowing (2007)

Web Customer Modeling for Automated Session Prioritization on High Traffic Sites (2007)

We present a new approach for dealing with distribution change and concept drift when learning from data sequences that may vary with time. We use sliding windows whose size, instead of being fixed a...

Nicolás Poggi, Toni Moreno, Josep Lluis Berral, Ricard Gavaldà, Jordi Torres

on high traffic sites

Learning from time-changing data with adaptive windowing (2007)

Kalman Filters and Adaptive Windows for Learning in Data Streams (2006)

We present a new approach for dealing with distribution change and concept drift when learning from data sequences that may vary with time. We use sliding windows whose size, instead of being fixed a...

Early Drift Detection Method (2006)

We study the combination of Kalman filter and a recently proposed algorithm for dynamically maintaining a sliding window, for learning from streams of examples. We integrate this idea into two well...

Baena-García, Manuel, Fidalgo, Raul, Bifet, Albert, Gavaldà, Ricard, Morales-Bueno, Rafael

An emerging problem in Data Streams is the detection of concept drift. This problem is aggravated when the drift is gradual over time. In this work we de¯ne a method for detecting concept drift,...

PAC-learning of Markov models with hidden state (2006)

Gavaldà, Ricard, Keller, Philipp W., Pineau, Joelle, Precup, Doina

The standard approach for learning Markov Models with Hidden State uses the Expectation-Maximization framework. While this approach had a significant impact on several practical applications (e.g....

Kalman filters and adaptive windows for learning in data streams (2006)

PAC-Learning of Markov Models with Hidden State (2006)

Abstract. We study the combination of Kalman filter and a recently proposed algorithm for dynamically maintaining a sliding window, for learning from streams of examples. We integrate this idea into...

Ricard Gavaldà, Joelle Pineau, Doina Precup

Abstract. The standard approach for learning Markov Models with Hidden State uses the Expectation-Maximization framework. While this approach had a significant impact on several practical...

Kalman filters and adaptive windows for learning in data streams (2006)

PAC-Learning of Markov Models with Hidden State (2006)

Abstract. We study the combination of Kalman filter and a recently proposed algorithm for dynamically maintaining a sliding window, for learning from streams of examples. We integrate this idea into...

Ricard Gavaldà, Joelle Pineau, Doina Precup

Abstract. The standard approach for learning Markov Models with Hidden State uses the Expectation-Maximization framework. While this approach had a significant impact on several practical...

Tractable Clones of Polynomials over Semigroups (2005)

Dalmau, Victor, Gavaldà, Ricard, Tesson, Pascal, Thérien, Denis

We contribute to the algebraic study of the complexity of constraint satisfaction problems. We give a new sufficient condition on a set of relations $\Gamma$ over a domain $S$ for the tractability of...

An algebraic view on exact learning from queries (2005)

Gavaldà, Ricard

We survey some recent work on the learnability of classes of functions defined by programs over monoids or semigroups in Angluin's model of exact learning from queries. This is ongoing work with D....

An Optimal Anytime Estimator Algorithm (2004)

Gavaldà, Ricard

In many applications a key step is estimating some unknown quantity X from a sequence of trials, each having expected value X. Optimal algorithms are known when the task is to estimate X within a...

Adaptive Sampling Methods for Scaling Up Knowledge Discovery Algorithms (1999)

Adaptive Sampling Methods for Scaling Up Knowledge Discovery Algorithms (1999)

. Scalability is a key requirement for any KDD and data mining algorithm, and one of the biggest research challenges is to develop methods that allow to use large amounts of data. One possible...

Adaptive Sampling Methods for Scaling Up Knowledge Discovery Algorithms (1999)

Scalability is a key requirement for any KDD and data mining algorithm, and one of the biggest research challenges is to develop methods that allow to use huge amount of data. One possible approach...

Practical Algorithms for On-line Sampling (1998)

Scalability is a key requirement for any KDD and data mining algorithm, and one of the biggest research challenges is to develop methods that allow to use large amounts of data. One possible approach...

Practical Algorithms for On-line Sampling (1998)

One of the core applications of machine learning to knowledge discovery consists on building a function (a hypothesis) from a given amount of data (for instance a decision tree or a neural network)...

Theoretical Analysis of Algorithms for On-line Selection (1998)

One of the core applications of machine learning to knowledge discovery consists on building a function (a hypothesis) from a given amount of data (for instance a decision tree or a neural network)...