Paolo Bientinesi

Jag är professor i datavetenskap med inriktning mot högprestandaberäkningar.

Vetenskaplig meritering: Docent

0000-0002-4972-7097

Kontakt

paolo.bientinesi@umu.se

072-085 61 80

Verksam vid

Professor vid Institutionen för datavetenskap

MIT-huset, Umeå universitet, UMIT, MIT.A.221 Umeå universitet, 901 87 Umeå

Anknuten som professor till Högpresterande beräkningscentrum Norr (HPC2N) Roll: Föreståndare

MIT huset, Campustorget 5, 2 tr Umeå universitet, 901 87 Umeå

Mina forskningsintressen berör numerisk linjär algebra, tensoroperationer, prestandamodellering och förutsägelse, datormusik och automatisk generering av algoritmer och kod.

Jag är särskilt intresserad av tvärvetenskaplig forskning, till exempel inom materialvetenskap, molekylär dynamik, såväl som beräkningskemi, biologi och fysik.

Den röda tråden i min forskning är utnyttjande av domänspecifik, problemspecifik och arkitekturspecifik kunskap för utveckling av högpresterande algoritmer.

Mer information om min forskning, CV och publikationer

In-depth performance analysis of the state-of-the-art algorithm for automatic drum transcription

25th International Society for Music Information Retrieval Conference (ISMIR), San Francisco, USA, 10-14 november, 2024

Zehren, Mickaël; Alunno, Marco; Bientinesi, Paolo

High-quality and reproducible automatic drum transcription from crowdsourced data

Signals, MDPI 2023, Vol. 4, (4) : 768-787

Zehren, Mickaël; Alunno, Marco; Bientinesi, Paolo

Work-Stealing Prefix Scan: Addressing Load Imbalance in Large-Scale Image Registration

IEEE Transactions on Parallel and Distributed Systems, IEEE Computer Society 2022, Vol. 33, (3) : 523-535

Copik, Marcin; Grosser, Tobias; Hoefler, Torsten; et al.

Editorial: High-performance tensor computations in scientific computing and data science

Frontiers in Applied Mathematics and Statistics, Frontiers Media S.A. 2022, Vol. 8

Di Napoli, Edoardo; Bientinesi, Paolo; Li, Jiajia; et al.

FLOPs as a discriminant for dense linear algebra algorithms

ICPP '22: proceedings of the 51st international conference on parallel processing

López, Francisco; Karlsson, Lars; Bientinesi, Paolo

The linear algebra mapping problem: Current state of linear algebra languages and libraries

ACM Transactions on Mathematical Software, ACM Digital Library 2022, Vol. 48, (3)

Psarras, Christos; Barthels, Henrik; Bientinesi, Paolo

Algorithm 1026: Concurrent Alternating Least Squares for Multiple Simultaneous Canonical Polyadic Decompositions

ACM Transactions on Mathematical Software, ACM Digital Library 2022, Vol. 48, (3)

Psarras, Christos; Karlsson, Lars; Bro, Rasmus; et al.

Accelerating Jackknife Resampling for the Canonical Polyadic Decomposition

Frontiers in Applied Mathematics and Statistics, Vol. 8

Psarras, Christos; Karlsson, Lars; Bro, Rasmus; et al.

Benchmarking the Linear Algebra Awareness of TensorFlow and PyTorch

Proceedings: 2022 IEEE 36th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2022, IEEE 2022 : 924-933

Sankaran, Aravind; Alashti, Navid Akbari; Psarras, Christos; et al.

A test for FLOPs as a discriminant for linear algebra algorithms

2022 IEEE 34th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), IEEE 2022 : 221-230

Sankaran, Aravind; Bientinesi, Paolo

Automatic detection of cue points for the emulation of DJ mixing

Computer music journal, MIT Press 2022, Vol. 46, (3) : 67-82

Zehren, Mickaël; Alunno, Marco; Bientinesi, Paolo

Linnea: Automatic Generation of Efficient Linear Algebra Programs

ACM Transactions on Mathematical Software, ACM Digital Library 2021, Vol. 47, (3)

Barthels, Henrik; Psarras, Christos; Bientinesi, Paolo

Rational spectral filters with optimal convergence rate

SIAM Journal on Scientific Computing, Society for Industrial and Applied Mathematics Publications 2021, Vol. 43, (4) : A2660-A2684

Kollnig, Konrad; Bientinesi, Paolo; Di Napoli, Edoardo A.

Performance Comparison for Scientific Computations on the Edge via Relative Performance

IEEE International Symposium on Parallel and Distributed Processing Workshops and Phd Forum (IPDPSW): 2021 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), IEEE 2021 : 887-895

Sankaran, Aravind; Bientinesi, Paolo

ADTOF: A large dataset of non-synthetic music for automatic drum transcription

Proceedings of the 22nd International Society for Music Information Retrieval Conference : 818-824

Zehren, Mickaël; Alunno, Marco; Bientinesi, Paolo

Automatic Generation of Efficient Linear Algebra Programs

PASC '20: Proceedings of the Platform for Advanced Scientific Computing Conference

Barthels, Henrik; Psarras, Christos; Bientinesi, Paolo

Accelerating Deep Learning Inference in Constrained Embedded Devices Using Hardware Loops and a Dot Product Unit

IEEE Access, IEEE 2020, Vol. 8 : 165913-165926

Vreča, Jure; Sturm, Karl J. X.; Gungl, Ernest; et al.

M-DJCUE: a manually annotated dataset of cue points

20th International Society for Music Information Retrieval Conference: Across the bridge, Delft, The Netherlands, November 4-8, 2019