Exploració per autor "Arnau Montañés, José María"
Ara es mostren els items 1-20 de 40
-
A low-power hardware accelerator for ORB feature extraction in self-driving cars
(Institute of Electrical and Electronics Engineers (IEEE), 2021)
Text en actes de congrés
Accés obertSimultaneous Localization And Mapping (SLAM) is a key component for autonomous navigation. SLAM consists of building and creating a map of an unknown environment while keeping track of the exploring agent's location within ... -
A low-power, high-performance speech recognition accelerator
(Institute of Electrical and Electronics Engineers (IEEE), 2019-12-01)
Article
Accés obertAutomatic Speech Recognition (ASR) is becoming increasingly ubiquitous, especially in the mobile segment. Fast and accurate ASR comes at high energy cost, not being affordable for the tiny power-budgeted mobile devices. ... -
A novel register renaming technique for out-of-order processors
(Institute of Electrical and Electronics Engineers (IEEE), 2018)
Text en actes de congrés
Accés restringit per política de l'editorialModern superscalar processors support a large number of in-flight instructions, which requires sizeable register files. Conventional register renaming techniques allocate a new storage location, i.e. physical register, for ... -
A programmable accelerator for streaming automatic speech recognition on edge devices
(2022)
Text en actes de congrés
Accés obertAutomatic Speech Recognition (ASR) is quickly becoming a mainstream technology, mainly driven by the outstanding accuracy achieved by modern systems based on machine learning. However, these systems often require billions ... -
An ultra low-power hardware accelerator for acoustic scoring in speech recognition
(Institute of Electrical and Electronics Engineers (IEEE), 2017)
Text en actes de congrés
Accés restringit per política de l'editorialAccurate, real-time Automatic Speech Recognition (ASR) comes at a high energy cost, so accuracy has often to be sacrificed in order to fit the strict power constraints of mobile systems. However, accuracy is extremely ... -
An ultra low-power hardware accelerator for automatic speech recognition
(IEEE Press, 2016)
Text en actes de congrés
Accés obertAutomatic Speech Recognition (ASR) is becoming increasingly ubiquitous, especially in the mobile segment. Fast and accurate ASR comes at a high energy cost which is not affordable for the tiny power budget of mobile devices. ... -
Boosting LSTM performance through dynamic precision selection
(Institute of Electrical and Electronics Engineers (IEEE), 2020)
Text en actes de congrés
Accés obertThe use of low numerical precision is a fundamental optimization included in modern accelerators for Deep Neural Networks (DNNs). The number of bits of the numerical representation is set to the minimum precision that is ... -
Boosting point cloud search with a vector unit
(2023)
Report de recerca
Accés obertModern robots collect and process point clouds to perform accurate registration and segmentation. The most time-consuming kernel within point cloud processing -namely neighbor search- relies on appropriate data structures, ... -
CGPA: Coarse-Grained Pruning of Activations for Energy-Efficient RNN Inference
(2019-09-01)
Article
Accés obertRecurrent neural networks (RNNs) perform element-wise multiplications across the activations of gates. We show that a significant percentage of activations are saturated and propose coarse-grained pruning of activations ... -
Characterizing self-driving tasks in general-purpose architectures
(European Network of Excellence on High Performance and Embedded Architecture and Compilation (HiPEAC), 2021-09-15)
Capítol de llibre
Accés obertAutonomous Vehicles (AVs) have the potential to radically change the automotive industry. How- ever, computing solutions for AVs have to meet severe performance constraints to guarantee a safe driving experience. Current ... -
Computation reuse in DNNs by exploiting input similarity
(Institute of Electrical and Electronics Engineers (IEEE), 2018)
Text en actes de congrés
Accés restringit per política de l'editorialIn recent years, Deep Neural Networks (DNNs) have achieved tremendous success for diverse problems such as classification and decision making. Efficient support for DNNs on CPUs, GPUs and accelerators has become a prolific ... -
CREW: Computation reuse and efficient weight storage for hardware-accelerated MLPs and RNNs
(2022-08-01)
Article
Accés obertDeep Neural Networks (DNNs) have achieved tremendous success for cognitive applications. The core operation in a DNN is the dot product between quantized inputs and weights. Prior works exploit the weight/input repetition ... -
Demystifying power and performance bottlenecks in autonomous driving systems
(Institute of Electrical and Electronics Engineers (IEEE), 2020)
Text en actes de congrés
Accés obertAutonomous Vehicles (AVs) have the potential to radically change the automotive industry. However, computing solutions for AVs have to meet severe performance and power constraints to guarantee a safe driving experience. ... -
Design and evaluation of an ultra low-power human-quality speech recognition system
(2020-11)
Article
Accés obertAutomatic Speech Recognition (ASR) has experienced a dramatic evolution since pioneer development of Bell Lab’s single-digit recognizer more than 50 years ago. Current ASR systems have taken advantage of the tremendous ... -
DNN pruning with principal component analysis and connection importance estimation
(2022-01)
Article
Accés obertDNN pruning reduces memory footprint and computational work of DNN-based solutions to improve performance and energy-efficiency. An effective pruning scheme should be able to systematically remove connections and/or neurons ... -
E-BATCH: Energy-efficient and high-throughput RNN batching
(2022-03)
Article
Accés obertRecurrent Neural Network (RNN) inference exhibits low hardware utilization due to the strict data dependencies across time-steps. Batching multiple requests can increase throughput. However, RNN batching requires a large ... -
E-PUR: an energy-efficient processing unit for recurrent neural networks
(2018)
Text en actes de congrés
Accés restringit per política de l'editorialRecurrent Neural Networks (RNNs) are a key technology for emerging applications such as automatic speech recognition, machine translation or image description. Long Short Term Memory (LSTM) networks are the most successful ... -
Eliminating redundant fragment shader executions on a mobile GPU via hardware memoization
(2014)
Text en actes de congrés
Accés restringit per política de l'editorialRedundancy is at the heart of graphical applications. In fact, generating an animation typically involves the succession of extremely similar images. In terms of rendering these images, this behavior translates into the ... -
Energy-efficient mobile GPU systems
(Universitat Politècnica de Catalunya, 2015-04-24)
Tesi
Accés obertThe design of mobile GPUs is all about saving energy. Smartphones and tablets are battery-operated and thus any type of rendering needs to use as little energy as possible. Furthermore, smartphones do not include sophisticated ... -
Energy-efficient stream compaction through filtering and coalescing accesses in GPGPU memory partitions
(Institute of Electrical and Electronics Engineers (IEEE), 2022-07-01)
Article
Accés obertGraph-based applications are essential in emerging domains such as data analytics or machine learning. Data gathering in a knowledge-based society requires great data processing efficiency. High-throughput GPGPU architectures ...
