Zobrazeno 1 - 10
of 328
pro vyhledávání: '"HAMDIOUI, SAID"'
Autor:
Huijbregts, Lucas, Hsiao-Hsuan, Liu, Detterer, Paul, Hamdioui, Said, Yousefzadeh, Amirreza, Bishnoi, Rajendra
Current Artificial Intelligence (AI) computation systems face challenges, primarily from the memory-wall issue, limiting overall system-level performance, especially for Edge devices with constrained battery budgets, such as smartphones, wearables, a
Externí odkaz:
http://arxiv.org/abs/2410.09130
Autor:
Dobrita, Alexandra, Yousefzadeh, Amirreza, Thorpe, Simon, Vadivel, Kanishkan, Detterer, Paul, Tang, Guangzhi, van Schaik, Gert-Jan, Konijnenburg, Mario, Gebregiorgis, Anteneh, Hamdioui, Said, Sifalakis, Manolis
For Edge AI applications, deploying online learning and adaptation on resource-constrained embedded devices can deal with fast sensor-generated streams of data in changing environments. However, since maintaining low-latency and power-efficient infer
Externí odkaz:
http://arxiv.org/abs/2406.17285
Autor:
Siddiqi, Muhammad Ali, Hernández, Jan Andrés Galvan, Gebregiorgis, Anteneh, Bishnoi, Rajendra, Strydis, Christos, Hamdioui, Said, Taouil, Mottaqiallah
Publikováno v:
Proceedings of the 2023 26th Euromicro Conference on Digital System Design (DSD)
Next-generation personalized healthcare devices are undergoing extreme miniaturization in order to improve user acceptability. However, such developments make it difficult to incorporate cryptographic primitives using available target technologies si
Externí odkaz:
http://arxiv.org/abs/2404.00125
Autor:
Shahroodi, Taha, Cardoso, Raphael, Wong, Stephan, Bosio, Alberto, O'Connor, Ian, Hamdioui, Said
State-of-the-Art (SotA) hardware implementations of Deep Neural Networks (DNNs) incur high latencies and costs. Binary Neural Networks (BNNs) are potential alternative solutions to realize faster implementations without losing accuracy. In this paper
Externí odkaz:
http://arxiv.org/abs/2401.17724
Autor:
Van Zegbroeck, Arne, Anagnostou, Pantazis, Hamdioui, Said, Adelmann, Christop, Ciubotaru, Florin, Cotofana, Sorin
While Spin Waves (SW) interaction provides natural support for low power Majority (MAJ) gate implementations many hurdles still exists on the road towards the realization of practically relevant SW circuits. In this paper we leave the SW interaction
Externí odkaz:
http://arxiv.org/abs/2401.12136
Autor:
Siddiqi, Muhammad Ali, Vrijenhoek, David, Landsmeer, Lennart P. L., van der Kleij, Job, Gebregiorgis, Anteneh, Romano, Vincenzo, Bishnoi, Rajendra, Hamdioui, Said, Strydis, Christos
Publikováno v:
21st ACM International Conference on Computing Frontiers Proceedings, 2024
Electrophysiological recordings of neural activity in a mouse's brain are very popular among neuroscientists for understanding brain function. One particular area of interest is acquiring recordings from the Purkinje cells in the cerebellum in order
Externí odkaz:
http://arxiv.org/abs/2311.04808
With the recent move towards sequencing of accurate long reads, finding solutions that support efficient analysis of these reads becomes more necessary. The long execution time required for sequence alignment of long reads negatively affects genomic
Externí odkaz:
http://arxiv.org/abs/2310.15634
Autor:
Shahroodi, Taha, Singh, Gagandeep, Zahedi, Mahdi, Mao, Haiyu, Lindegger, Joel, Firtina, Can, Wong, Stephan, Mutlu, Onur, Hamdioui, Said
Basecalling, an essential step in many genome analysis studies, relies on large Deep Neural Networks (DNNs) to achieve high accuracy. Unfortunately, these DNNs are computationally slow and inefficient, leading to considerable delays and resource cons
Externí odkaz:
http://arxiv.org/abs/2310.04366
Autor:
Gursoy, Cemil Cem, Kraak, Daniel, Ahmed, Foisal, Taouil, Mottaqiallah, Jenihhin, Maksim, Hamdioui, Said
Memory designs require timing margins to compensate for aging and fabrication process variations. With technology downscaling, aging mechanisms became more apparent, and larger margins are considered necessary. This, in return, means a larger area re
Externí odkaz:
http://arxiv.org/abs/2212.09356
Applications of Binary Neural Networks (BNNs) are promising for embedded systems with hard constraints on computing power. Contrary to conventional neural networks with the floating-point datatype, BNNs use binarized weights and activations which add
Externí odkaz:
http://arxiv.org/abs/2211.06261