Zobrazeno 1 - 10
of 2 149
pro vyhledávání: '"Haghani, P."'
Boosting Sample Efficiency and Generalization in Multi-agent Reinforcement Learning via Equivariance
Multi-Agent Reinforcement Learning (MARL) struggles with sample inefficiency and poor generalization [1]. These challenges are partially due to a lack of structure or inductive bias in the neural networks typically used in learning the policy. One su
Externí odkaz:
http://arxiv.org/abs/2410.02581
Autor:
Aliannejadi, R., Haghani, Z.
Publikováno v:
IJAA. 11 (2024) 45-57
We investigate the physical properties of quark stars with two different equations of state in generalized hybrid metric-Palatini gravity. This theory corresponds to a bi-scalar gravitational theory in which there are two non-minimally-coupled scalar
Externí odkaz:
http://arxiv.org/abs/2410.00159
Autor:
Gaur, Neeraj, Agrawal, Rohan, Wang, Gary, Haghani, Parisa, Rosenberg, Andrew, Ramabhadran, Bhuvana
This paper introduces ASTRA, a novel method for improving Automatic Speech Recognition (ASR) through text injection.Unlike prevailing techniques, ASTRA eliminates the need for sampling to match sequence lengths between speech and text modalities. Ins
Externí odkaz:
http://arxiv.org/abs/2406.06664
Autor:
Haghani, Milad, Ronchi, Enrico
The paper "Simulating dynamical features of escape panic" by Helbing, Farkas, and Vicsek, published over two decades ago in Nature, has left an indelible mark on the field of crowd dynamics. With nearly 3,000 citations to date, according to the Web o
Externí odkaz:
http://arxiv.org/abs/2310.20506
Autor:
Ngai, Hillary, Agrawal, Rohan, Gaur, Neeraj, Huang, Ronny, Haghani, Parisa, Mengibar, Pedro Moreno
Adapters are an efficient, composable alternative to full fine-tuning of pre-trained models and help scale the deployment of large ASR models to many tasks. In practice, a task ID is commonly prepended to the input during inference to route to single
Externí odkaz:
http://arxiv.org/abs/2310.13015
Publikováno v:
Safety Science 172C (2024) 106423
The portrayal of crowd accidents by the media can influence public understanding and emotional response, shaping societal perceptions and potentially impacting safety measures and preparedness strategies. This paper critically examines the portrayal
Externí odkaz:
http://arxiv.org/abs/2309.14633
Autor:
Blau, Yochai, Agrawal, Rohan, Madmony, Lior, Wang, Gary, Rosenberg, Andrew, Chen, Zhehuai, Gekhman, Zorik, Beryozkin, Genady, Haghani, Parisa, Ramabhadran, Bhuvana
Accurate recognition of specific categories, such as persons' names, dates or other identifiers is critical in many Automatic Speech Recognition (ASR) applications. As these categories represent personal information, ethical use of this data includin
Externí odkaz:
http://arxiv.org/abs/2308.07393
This paper presents a state-of-the-art model for transcribing speech in any language into the International Phonetic Alphabet (IPA). Transcription of spoken languages into IPA is an essential yet time-consuming process in language documentation, and
Externí odkaz:
http://arxiv.org/abs/2308.03917
Autor:
Osman Sharifi, Viktoria Haghani, Kari E. Neier, Keith J. Fraga, Ian Korf, Sophia M. Hakam, Gerald Quon, Nelson Johansen, Dag H. Yasui, Janine M. LaSalle
Publikováno v:
Communications Biology, Vol 7, Iss 1, Pp 1-14 (2024)
Abstract Dominant X-linked diseases are uncommon due to female X chromosome inactivation (XCI). While random XCI usually protects females against X-linked mutations, Rett syndrome (RTT) is a female neurodevelopmental disorder caused by heterozygous M
Externí odkaz:
https://doaj.org/article/27f2deea1c084c859913e45acdf66dda
Autor:
Etrat Hooshmandi, Somayeh Akbari, Negin Gharbi, Mojtaba Ghobadi, Ava Shahrokhabadi, Mohammad Saied Salehi, Afsoon Afshari, Masoud Haghani, Mahnaz Bayat
Publikováno v:
Egyptian Journal of Medical Human Genetics, Vol 25, Iss 1, Pp 1-5 (2024)
Abstract Background Dementia, characterized by synaptic dysfunction and memory loss, presents a significant challenge in medical research. Long non-coding RNAs (lncRNAs), known for their crucial roles in regulating gene expression, represent a gap in
Externí odkaz:
https://doaj.org/article/a0d27f02ece8480da2b6a6fa9eb5f412