Zobrazeno 1 - 10
of 4 173
pro vyhledávání: '"Mohajer, A. A."'
Second-order methods are widely adopted to improve the convergence rate of learning algorithms. In federated learning (FL), these methods require the clients to share their local Hessian matrices with the parameter server (PS), which comes at a prohi
Externí odkaz:
http://arxiv.org/abs/2412.03867
Autor:
Mohajer, Mohammadreza
In this thesis, we aim to develop p-adic analogs of known results for classical periods, focusing specifically on 1-motives. We establish an integration theory for 1-motives with good reductions, which generalizes the Colmez-Fontaine-Messing p-adic i
Externí odkaz:
http://arxiv.org/abs/2411.03118
Autor:
Mohajer, Mohammadreza Bakhshizadeh, Barletta, Luca, Tuninetti, Daniela, Tomasoni, Alessandro, Iacono, Daniele Lo, Osnato, Fabio
Integrated sensing and communication is regarded as a key enabler for next-generation wireless networks. To optimize the transmitted waveform for both sensing and communication, various performance metrics must be considered. This work focuses on sen
Externí odkaz:
http://arxiv.org/abs/2410.22941
Autor:
Aleithan, Reem, Xue, Haoran, Mohajer, Mohammad Mahdi, Nnorom, Elijah, Uddin, Gias, Wang, Song
Large Language Models (LLMs) in Software Engineering (SE) can offer assistance for coding. To facilitate a rigorous evaluation of LLMs in practical coding contexts, Carlos et al. introduced the SWE-bench dataset, which comprises 2,294 real-world GitH
Externí odkaz:
http://arxiv.org/abs/2410.06992
Autor:
Harzevili, Nima Shiri, Mohajer, Mohammad Mahdi, Shin, Jiho, Wei, Moshi, Uddin, Gias, Yang, Jinqiu, Wang, Junjie, Wang, Song, Ming, Zhen, Jiang, Nagappan, Nachiappan
Checker bugs in Deep Learning (DL) libraries are critical yet not well-explored. These bugs are often concealed in the input validation and error-checking code of DL libraries and can lead to silent failures, incorrect results, or unexpected program
Externí odkaz:
http://arxiv.org/abs/2410.06440
Autor:
Shahandashti, Kimya Khakzad, Mohajer, Mohammad Mahdi, Belle, Alvine Boaye, Wang, Song, Hemmati, Hadi
Program slicing is a critical technique in software engineering, enabling developers to isolate relevant portions of code for tasks such as bug detection, code comprehension, and debugging. In this study, we investigate the application of large langu
Externí odkaz:
http://arxiv.org/abs/2409.12369
Autor:
Hamidi, Shayan Mohajer, Bereyhi, Ali
Quantization is a common approach to mitigate the communication cost of federated learning (FL). In practice, the quantized local parameters are further encoded via an entropy coding technique, such as Huffman coding, for efficient data compression.
Externí odkaz:
http://arxiv.org/abs/2409.06319
Autor:
Mohajer, Abolfazl
Using the theory of Higgs bundles and their stabitlity properties associated to fibered surfaces and the Viehweg-Zuo characterization of Shimura curves in the moduli space of abelian varieties in terms of Higgs bundles, we prove that there does not e
Externí odkaz:
http://arxiv.org/abs/2408.10219
Recently, it was shown that the role of the teacher in knowledge distillation (KD) is to provide the student with an estimate of the true Bayes conditional probability density (BCPD). Notably, the new findings propose that the student's error rate ca
Externí odkaz:
http://arxiv.org/abs/2407.18041
Human Activity Recognition (HAR) stands as a pivotal technique within pattern recognition, dedicated to deciphering human movements and actions utilizing one or multiple sensory inputs. Its significance extends across diverse applications, encompassi
Externí odkaz:
http://arxiv.org/abs/2406.11164