Browsing by Author "Buchaca Prats, David"

Now showing items 1-10 of 10

    • A highly parameterizable framework for Conditional Restricted Boltzmann Machine based workloads accelerated with FPGAs and OpenCL 

      Jaksic, Zoran; Cadenelli, Nicola; Buchaca Prats, David; Polo Bardés, Jordà; Berral García, Josep Lluís; Carrera Pérez, David (Elsevier, 2020-03-01)
      Article
      Open Access
      Conditional Restricted Boltzmann Machine (CRBM) is a promising candidate for a multidimensional system modeling that can learn a probability distribution over a set of data. It is a specific type of an artificial neural ...

    • A multilayer extension of the similarity neural network 

      Buchaca Prats, David (Universitat Politècnica de Catalunya, 2014-11-24)
      Master thesis
      Open Access
      Aquest projecte ajunta idees de les radial basis functions, i el multilayer perceptron per a desenvolupar una altra arquitectura de xarxa neuronal artificial i un mètode per a poder-la entrenar. És una extensió de la ...

    • Automatic Generation of Workload Profiles Using Unsupervised Learning Pipelines 

      Buchaca Prats, David; Berral, Josep Ll.; Carrera, David (IEEE, 2018-03)
      Article
      Open Access
      The complexity of resource usage and power consumption on cloud-based applications makes the understanding of application behavior through expert examination difficult. The difficulty increases when applications are seen ...

    • Automatic generation of workload profiles using unsupervised learning pipelines 

      Buchaca Prats, David; Bernal García, Josep Lluis; Carrera Pérez, David (2017-12-27)
      Article
      Open Access
      The complexity of resource usage and power consumption on cloud-based applications makes the understanding of application behavior through expert examination difficult. The difficulty increases when applications are seen ...

    • Improving maritime traffic emission estimations on missing data with CRBMs 

      Gutiérrez Torre, Alberto; Berral García, Josep Lluís; Buchaca Prats, David; Guevara Vilardell, Marc; Soret, Albert; Carrera Pérez, David (Elsevier, 2020-07-07)
      Article
      Restricted access - publisher's policy
      Maritime traffic emissions are a major concern to governments as they heavily impact the Air Quality in coastal cities. Ships use the Automatic Identification System (AIS) to continuously report position and speed among ...

    • Learning workload behaviour models from monitored time-series for resource estimation towards data center optimization 

      Buchaca Prats, David (Universitat Politècnica de Catalunya, 2021-01-14)
      Doctoral thesis
      Open Access
      In recent years there has been an extraordinary growth of the demand of Cloud Computing resources executed in Data Centers. Modern Data Centers are complex systems that need management. As distributed computing systems ...

    • Proactive container auto-scaling for cloud native machine learning services 

      Buchaca Prats, David; Berral García, Josep Lluís; Wang, Chen; Youssef, Alaa (Institute of Electrical and Electronics Engineers (IEEE), 2020)
      Conference lecture
      Open Access
      Understanding the resource usage behaviors of the ever-increasing machine learning workloads are critical to cloud providers offering Machine Learning (ML) services. Capable of auto-scaling resources for customer workloads ...

    • Stopping criteria in contrastive divergence: Alternatives to the reconstruction error 

      Buchaca Prats, David; Romero Merino, Enrique; Mazzanti Castrillejo, Fernando Pablo; Delgado Pin, Jordi (2014)
      Conference report
      Open Access
      Restricted Boltzmann Machines (RBMs) are general unsupervised learning devices to ascertain generative models of data distributions. RBMs are often trained using the Contrastive Divergence learning algorithm (CD), an ...

    • Weighted contrastive divergence 

      Romero Merino, Enrique; Mazzanti Castrillejo, Fernando Pablo; Delgado Pin, Jordi; Buchaca Prats, David (2019-06)
      Article
      Open Access
      Learning algorithms for energy based Boltzmann architectures that rely on gradient descent are in general computationally prohibitive, typically due to the exponential number of terms involved in computing the partition ...

    • You only run once: Spark auto-tuning from a single run 

      Buchaca Prats, David; Albuquerque Portella, Felipe; Costa, Carlos H. A.; Berral García, Josep Lluís (2020-12)
      Article
      Open Access
      Tuning configurations of Spark jobs is not a trivial task. State-of-the-art auto-tuning systems are based on iteratively running workloads with different configurations. During the optimization process, the relevant features ...