Departament d'Arquitectura de Computadors
Enviaments recents
-
Efficient hardware acceleration of deep neural networks via arithmetic complexity reduction
(Universitat Politècnica de Catalunya, 2023-10-26)
Tesi
Accés obert(English) Over the past decade, significant progresses in the field of artificial intelligence have led to remarkable advancements in a wide range of technologies. Deep learning, a subfield of machine learning centered ... -
Acceleration of automatic speech recognition for low-power devices
(Universitat Politècnica de Catalunya, 2022-11-09)
Tesi
Accés obert(English) In this thesis, we study the challenges preventing ASR deployment on edge devices and propose innovations to tackle them, hopefully moving the technology a step forward to the future. First, we characterize ... -
Job scheduling for disaggregated memory in high performance computing systems
(Universitat Politècnica de Catalunya, 2023-10-09)
Tesi
Accés obert(English) In a typical HPC cluster system, a node is the elemental component unit of this architecture. Memory and compute resources are tightly coupled in each node and the rigid boundaries between nodes limits compute ... -
Deep learning for spatio-temporal forecasting: benchmarks, methods, and insights from mobility and weather predictions
(Universitat Politècnica de Catalunya, 2023-10-30)
Tesi
Accés obert(English) This thesis explores the intersection of deep learning and spatio-temporal forecasting, focusing on the challenges and opportunities present in applying machine learning methods to predict complex geospatial and ... -
Adapting floating-point precision to accelerate deep neural network training
(Universitat Politècnica de Catalunya, 2023-10-18)
Tesi
Accés obert(English) Deep Neural Networks (DNNs) have become ubiquitous in a wide range of application domains. Despite their success, training DNNs is an expensive task which has motivated the use of reduced numerical precision ... -
Advanced hardware prefetching in virtual memory systems
(Universitat Politècnica de Catalunya, 2023-09-12)
Tesi
Accés obert(English) Despite groundbreaking technological innovations, the disparity between processor and memory speeds (known as Memory Wall) is still a major performance obstacle for modern systems. Hardware prefetching is a ... -
Software diagnostics for autonomous safety-critical control-systems based on artificial intelligence
(Universitat Politècnica de Catalunya, 2023-07-18)
Tesi
Accés obert(English) Machine Learning (ML) systems allow the efficient implementation of functionalities that can be hard to program by traditional software due to the high spectrum of inputs that hinder the definition of a specific ... -
Convergence of high performance computing, big data, and machine learning applications on containerized infrastructures
(Universitat Politècnica de Catalunya, 2023-07-17)
Tesi
Accés obert(English) The convergence of High Performance Computing (HPC), Big Data (BD), and Machine Learning (ML) in the computing continuum is being pursued in earnest across the academic and industry. We envision virtualization ... -
Scaling deep learning workloads. Applications in computer vision and seismology
(Universitat Politècnica de Catalunya, 2023-07-06)
Tesi
Accés obert(English) Deep learning techniques have an enormous impact on the state-of-the-art in many fields, such as computer vision, natural language processing, audio analysis and synthesis, and many others. The increasing computing ... -
Hardware/software solutions to enable the use of high-performance processors in the most stringent safety-critical systems
(Universitat Politècnica de Catalunya, 2023-07-19)
Tesi
Accés obert(English) Future Safety-Critical Systems require a boost in guaranteed performance in order to satisfy the increasing performance demands of the state-of-the-art complex software features. Ar1 approach to achieve these ... -
Leveraging graph neural networks for optimization and traffic compression in network digital twins
(Universitat Politècnica de Catalunya, 2023-07-17)
Tesi
Accés obert(English) In recent years, several industry sectors have adapted the Digital Twin (DT) paradigm to improve the performance of physical systems. This paradigm consists of leveraging computational methods to build high-fidelity ... -
Artificial intelligence solutions for quantum communications
(Universitat Politècnica de Catalunya, 2023-06-16)
Tesi
Accés obert(English) This Ph.D. thesis focuses on the application of intelligent models to Discrete-Variable Quantum Key Distribution (DV-QKD) protocol. The first objective focuses on providing a method for AI-based polarization ...